Standardized mortality
Observational study design strengths and weaknesses.
Very inexpensive Fast Easy to assign exposure levels | Inaccuracy of data Inability to control for confounders Difficulty identifying or quantifying denominator No demonstrated temporality | |
Very inexpensive Fast Outcome (death) well captured | Utilize deaths only Inaccuracy of data (death certificates) Inability to control for confounders | |
Reduces some types of bias Good for acute health outcomes with a defined exposure Cases act as their own control | Selection of comparison time point difficult Challenging to execute Prone to recall bias No demonstrated temporality | |
Inexpensive Timely Individualized data Ability to control for multiple confounders Can assess multiple outcomes | No temporality Not good for rare diseases Poor for diseases of short duration No demonstrated temporality | |
Inexpensive Timely Individualized data Ability to control for multiple confounders Good for rare diseases Can assess multiple exposures | Cannot calculate prevalence Can only assess one outcome Poor selection of controls can introduce bias May be difficult to identify enough cases Prone to recall bias No demonstrated temporality | |
Temporality demonstrated Individualized data Ability to control for multiple confounders Can assess multiple exposures Can assess multiple outcomes | Expensive Time intensive Not good for rare diseases |
Ecological study design.
The most basic observational study is an ecological study. This study design compares clusters of people, usually grouped based on their geographical location or temporal associations ( 1 , 2 , 6 , 9 ). Ecological studies assign one exposure level for each distinct group and can provide a rough estimation of prevalence of disease within a population. Ecological studies are generally retrospective. An example of an ecological study is the comparison of the prevalence of obesity in the United States and France. The geographic area is considered the exposure and the outcome is obesity. There are inherent potential weaknesses with this approach, including loss of data resolution and potential misclassification ( 10 , 11 , 13 , 18 , 19 ). This type of study design also has additional weaknesses. Typically these studies derive their data from large databases that are created for purposes other than research, which may introduce error or misclassification ( 10 , 11 ). Quantification of both the number of cases and the total population can be difficult, leading to error or bias. Lastly, due to the limited amount of data available, it is difficult to control for other factors that may mask or falsely suggest a relationship between the exposure and the outcome. However, ecological studies are generally very cost effective and are a starting point for hypothesis generation.
Proportional mortality ratio studies (PMR) utilize the defined well recorded outcome of death and subsequent records that are maintained regarding the decedent ( 1 , 6 , 8 , 20 ). By using records, this study design is able to identify potential relationships between exposures, such as geographic location, occupation, or age and cause of death. The epidemiological outcomes of this study design are proportional mortality ratio and standardized mortality ratio. In general these are the ratio of the proportion of cause-specific deaths out of all deaths between exposure categories ( 20 ). As an example, these studies can address questions about higher proportion of cardiovascular deaths among different ethnic and racial groups ( 21 ). A significant drawback to the PMR study design is that these studies are limited to death as an outcome ( 3 , 5 , 22 ). Additionally, the reliance on death records makes it difficult to control for individual confounding factors, variables that either conceal or falsely demonstrate associations between the exposure and outcome. An example of a confounder is tobacco use confounding the relationship between coffee intake and cardiovascular disease. Historically people often smoked and drank coffee while on coffee breaks. If researchers ignore smoking they would inaccurately find a strong relationship between coffee use and cardiovascular disease, where some of the risk is actually due to smoking. There are also concerns regarding the accuracy of death certificate data. Strengths of the study design include the well-defined outcome of death, the relative ease and low cost of obtaining data, and the uniformity of collection of these data across different geographical areas.
Cross-sectional studies are also called prevalence studies because one of the main measures available is study population prevalence ( 1 – 12 ). These studies consist of assessing a population, as represented by the study sample, at a single point in time. A common cross-sectional study type is the diagnostic accuracy study, which is discussed later. Cross-sectional study samples are selected based on their exposure status, without regard for their outcome status. Outcome status is obtained after participants are enrolled. Ideally, a wider distribution of exposure will allow for a higher likelihood of finding an association between the exposure and outcome if one exists ( 1 – 3 , 5 , 8 ). Cross-sectional studies are retrospective in nature. An example of a cross-sectional study would be enrolling participants who are either current smokers or never smokers, and assessing whether or not they have respiratory deficiencies. Random sampling of the population being assessed is more important in cross-sectional studies as compared to other observational study designs. Selection bias from non-random sampling may result in flawed measure of prevalence and calculation of risk. The study sample is assessed for both exposure and outcome at a single point in time. Because both exposure and outcome are assessed at the same time, temporality cannot be demonstrated, i.e. it cannot be demonstrated that the exposure preceded the disease ( 1 – 3 , 5 , 8 ). Point prevalence and period prevalence can be calculated in cross-sectional studies. Measures of risk for the exposure-outcome relationship that can be calculated in cross-sectional study design are odds ratio, prevalence odds ratio, prevalence ratio, and prevalence difference. Cross-sectional studies are relatively inexpensive and have data collected on an individual which allows for more complete control for confounding. Additionally, cross-sectional studies allow for multiple outcomes to be assessed simultaneously.
Case-control studies were traditionally referred to as retrospective studies, due to the nature of the study design and execution ( 1 – 12 , 23 , 24 ). In this study design, researchers identify study participants based on their case status, i.e. diseased or not diseased. Quantification of the number of individuals among the cases and the controls who are exposed allow for statistical associations between exposure and outcomes to be established ( 1 – 3 , 5 , 8 ). An example of a case control study is analysing the relationship between obesity and knee replacement surgery. Cases are participants who have had knee surgery, and controls are a random sampling of those who have not, and the comparison is the relative odds of being obese if you have knee surgery as compared to those that do not. Matching on one or more potential confounders allows for minimization of those factors as potential confounders in the exposure-outcome relationship ( 1 – 3 , 5 , 8 ). Additionally, case-control studies are at increased risk for bias, particularly recall bias, due to the known case status of study participants ( 1 – 3 , 5 , 8 ). Other points of consideration that have specific weight in case-control studies include the appropriate selection of controls that balance generalizability and minimize bias, the minimization of survivor bias, and the potential for length time bias ( 25 ). The largest strength of case-control studies is that this study design is the most efficient study design for rare diseases. Additional strengths include low cost, relatively fast execution compared to cohort studies, the ability to collect individual participant specific data, the ability to control for multiple confounders, and the ability to assess multiple exposures of interest. The measure of risk that is calculated in case-control studies is the odds ratio, which are the odds of having the exposure if you have the disease. Other measures of risk are not applicable to case-control studies. Any measure of prevalence and associated measures, such as prevalence odds ratio, in a case-control study is artificial because the researcher arbitrarily sets the proportion of cases to non-cases in this study design. Temporality can be suggested, however, it is rarely definitively demonstrated because it is unknown if the development of the disease truly preceded the exposure. It should be noted that for certain outcomes, particularly death, the criteria for demonstrating temporality in that specific exposure-outcome relationship are met and the use of relative risk as a measure of risk may be justified.
A case-crossover study relies upon an individual to act as their own control for comparison issues, thereby minimizing some potential confounders ( 1 , 5 , 12 ). This study design should not be confused with a crossover study design which is an interventional study type and is described below. For case-crossover studies, cases are assessed for their exposure status immediately prior to the time they became a case, and then compared to their own exposure at a prior point where they didn’t become a case. The selection of the prior point for comparison issues is often chosen at random or relies upon a mean measure of exposure over time. Case-crossover studies are always retrospective. An example of a case-crossover study would be evaluating the exposure of talking on a cell phone and being involved in an automobile crash. Cases are drivers involved in a crash and the comparison is that same driver at a random timeframe where they were not involved in a crash. These types of studies are particularly good for exposure-outcome relationships where the outcome is acute and well defined, e.g. electrocutions, lacerations, automobile crashes, etc. ( 1 , 5 ). Exposure-outcome relationships that are assessed using case-crossover designs should have health outcomes that do not have a subclinical or undiagnosed period prior to becoming a “case” in the study ( 12 ). The exposure is cell phone use during the exposure periods, both before the crash and during the control period. Additionally, the reliance upon prior exposure time requires that the exposure not have an additive or cumulative effect over time ( 1 , 5 ). Case-crossover study designs are at higher risk for having recall bias as compared with other study designs ( 12 ). Study participants are more likely to remember an exposure prior to becoming a case, as compared to not becoming a case.
Cohort studies involve identifying study participants based on their exposure status and either following them through time to identify which participants develop the outcome(s) of interest, or look back at data that were created in the past, prior to the development of the outcome. Prospective cohort studies are considered the gold standard of observational research ( 1 – 3 , 5 , 8 , 10 , 11 ). These studies begin with a cross-sectional study to categorize exposure and identify cases at baseline. Disease-free participants are then followed and cases are measured as they develop. Retrospective cohort studies also begin with a cross-sectional study to categorize exposure and identify cases. Exposures are then measured based on records created at that time. Additionally, in an ideal retrospective cohort, case status is also tracked using historical data that were created at that point in time. Occupational groups, particularly those that have regular surveillance or certifications such as Commercial Truck Drivers, are particularly well positioned for retrospective cohort studies because records of both exposure and outcome are created as part of commercial and regulatory purposes ( 8 ). These types of studies have the ability to demonstrate temporality and therefore identify true risk factors, not associated factors, as can be done in other types of studies.
Cohort studies are the only observational study that can calculate incidence, both cumulative incidence and an incidence rate ( 1 , 3 , 5 , 6 , 10 , 11 ). Also, because the inception of a cohort study is identical to a cross-sectional study, both point prevalence and period prevalence can be calculated. There are many measures of risk that can be calculated from cohort study data. Again, the measures of risk for the exposure-outcome relationship that can be calculated in cross-sectional study design of odds ratio, prevalence odds ratio, prevalence ratio, and prevalence difference can be calculated in cohort studies as well. Measures of risk that leverage a cohort study’s ability to calculate incidence include incidence rate ratio, relative risk, risk ratio, and hazard ratio. These measures that demonstrate temporality are considered stronger measures for demonstrating causation and identification of risk factors.
A specific study design is the diagnostic accuracy study, which is often used as part of the clinical decision making process. Diagnostic accuracy study designs are those that compare a new diagnostic method with the current “gold standard” diagnostic procedure in a cross-section of both diseased and healthy study participants. Gold standard diagnostic procedures are the current best-practice for diagnosing a disease. An example is comparing a new rapid test for a cancer with the gold standard method of biopsy. There are many intricacies to diagnostic testing study designs that should be considered. The proper selection of the gold standard evaluation is important for defining the true measures of accuracy for the new diagnostic procedure. Evaluations of diagnostic test results should be blinded to the case status of the participant. Similar to the intention-to-treat concept discussed later in interventional studies, diagnostic tests have a procedure of analyses called intention to diagnose (ITD), where participants are analysed in the diagnostic category they were assigned, regardless of the process in which a diagnosis was obtained. Performing analyses according to an a priori defined protocol, called per protocol analyses (PP or PPA), is another potential strength to diagnostic study testing. Many measures of the new diagnostic procedure, including accuracy, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio can be calculated. These measures of the diagnostic test allow for comparison with other diagnostic tests and aid the clinician in determining which test to utilize.
Interventional study designs, also called experimental study designs, are those where the researcher intervenes at some point throughout the study. The most common and strongest interventional study design is a randomized controlled trial, however, there are other interventional study designs, including pre-post study design, non-randomized controlled trials, and quasi-experiments ( 1 , 5 , 13 ). Experimental studies are used to evaluate study questions related to either therapeutic agents or prevention. Therapeutic agents can include prophylactic agents, treatments, surgical approaches, or diagnostic tests. Prevention can include changes to protective equipment, engineering controls, management, policy or any element that should be evaluated as to a potential cause of disease or injury.
A pre-post study measures the occurrence of an outcome before and again after a particular intervention is implemented. A good example is comparing deaths from motor vehicle crashes before and after the enforcement of a seat-belt law. Pre-post studies may be single arm, one group measured before the intervention and again after the intervention, or multiple arms, where there is a comparison between groups. Often there is an arm where there is no intervention. The no-intervention arm acts as the control group in a multi-arm pre-post study. These studies have the strength of temporality to be able to suggest that the outcome is impacted by the intervention, however, pre-post studies do not have control over other elements that are also changing at the same time as the intervention is implemented. Therefore, changes in disease occurrence during the study period cannot be fully attributed to the specific intervention. Outcomes measured for pre-post intervention studies may be binary health outcomes such as incidence or prevalence, or mean values of a continuous outcome such as systolic blood pressure may also be used. The analytic methods of pre-post studies depend on the outcome being measured. If there are multiple treatment arms, it is also likely that the difference from beginning to end within each treatment arm are analysed.
Non-randomized trials are interventional study designs that compare a group where an intervention was performed with a group where there was no intervention. These are convenient study designs that are most often performed prospectively and can suggest possible relationships between the intervention and the outcome. However, these study designs are often subject to many types of bias and error and are not considered a strong study design.
Randomized controlled trials (RCTs) are the most common type of interventional study, and can have many modifications ( 26 – 28 ). These trials take a homogenous group of study participants and randomly divide them into two separate groups. If the randomization is successful then these two groups should be the same in all respects, both measured confounders and unmeasured factors. The intervention is then implemented in one group and not the other and comparisons of intervention efficacy between the two groups are analysed. Theoretically, the only difference between the two groups through the entire study is the intervention. An excellent example is the intervention of a new medication to treat a specific disease among a group of patients. This randomization process is arguably the largest strength of an RCT ( 26 – 28 ). Additional methodological elements are utilized among RCTs to further strengthen the causal implication of the intervention’s impact. These include allocation concealment, blinding, measuring compliance, controlling for co-interventions, measuring dropout, analysing results by intention to treat, and assessing each treatment arm at the same time point in the same manner.
A crossover RCT is a type of interventional study design where study participants intentionally “crossover” to the other treatment arm. This should not be confused with the observational case-crossover design. A crossover RCT begins the same as a traditional RCT, however, after the end of the first treatment phase, each participant is re-allocated to the other treatment arm. There is often a wash-out period in between treatment periods. This design has many strengths, including demonstrating reversibility, compensating for unsuccessful randomization, and improving study efficiency by not using time to recruit subjects.
Allocation concealment theoretically guarantees that the implementation of the randomization is free from bias. This is done by ensuring that the randomization scheme is concealed from all individuals involved ( 26 – 30 ). A third party who is not involved in the treatment or assessment of the trial creates the randomization schema and study participants are randomized according to that schema. By concealing the schema, there is a minimization of potential deviation from that randomization, either consciously or otherwise by the participant, researcher, provider, or assessor. The traditional method of allocation concealment relies upon sequentially numbered opaque envelopes with the treatment allocation inside. These envelopes are generated before the study begins using the selected randomization scheme. Participants are then allocated to the specific intervention arm in the pre-determined order dictated by the schema. If allocation concealment is not utilized, there is the possibility of selective enrolment into an intervention arm, potentially with the outcome of biased results.
Blinding in an RCT is withholding the treatment arm from individuals involved in the study. This can be done through use of placebo pills, deactivated treatment modalities, or sham therapy. Sham therapy is a comparison procedure or treatment which is identical to the investigational intervention except it omits a key therapeutic element, thus rendering the treatment ineffective. An example is a sham cortisone injection, where saline solution of the same volume is injected instead of cortisone. This helps ensure that patients do not know if they are receiving the active or control treatment. The process of blinding is utilized to help ensure equal treatment of the different groups, therefore continuing to isolate the difference in outcome between groups to only the intervention being administered ( 28 – 31 ). Blinding within an RCT includes patient blinding, provider blinding, or assessor blinding. In some situations it is difficult or impossible to blind one or more of the parties involved, but an ideal study would have all parties blinded until the end of the study ( 26 – 28 , 31 , 32 ).
Compliance is the degree of how well study participants adhere to the prescribed intervention. Compliance or non-compliance to the intervention can have a significant impact on the results of the study ( 26 – 29 ). If there is a differentiation in the compliance between intervention arms, that differential can mask true differences, or erroneously conclude that there are differences between the groups when one does not exist. The measurement of compliance in studies addresses the potential for differences observed in intervention arms due to intervention adherence, and can allow for partial control of differences either through post hoc stratification or statistical adjustment.
Co-interventions, interventions that impact the outcome other than the primary intervention of the study, can also allow for erroneous conclusions in clinical trials ( 26 – 28 ). If there are differences between treatment arms in the amount or type of additional therapeutic elements then the study conclusions may be incorrect ( 29 ). For example, if a placebo treatment arm utilizes more over-the-counter medication than the experimental treatment arm, both treatment arms may have the same therapeutic improvement and show no effect of the experimental treatment. However, the placebo arm improvement is due to the over-the-counter medication and if that was prohibited, there may be a therapeutic difference between the two treatment arms. The exclusion or tracking and statistical adjustment of co-interventions serves to strengthen an RCT by minimizing this potential effect.
Participants drop out of a study for multiple reasons, but if there are differential dropout rates between intervention arms or high overall dropout rates, there may be biased data or erroneous study conclusions ( 26 – 28 ). A commonly accepted dropout rate is 20% however, studies with dropout rates below 20% may have erroneous conclusions ( 29 ). Common methods for minimizing dropout include incentivizing study participation or short study duration, however, these may also lead to lack of generalizability or validity.
Intention-to-treat (ITT) analysis is a method of analysis that quantitatively addresses deviations from random allocation ( 26 – 28 ). This method analyses individuals based on their allocated intervention, regardless of whether or not that intervention was actually received due to protocol deviations, compliance concerns or subsequent withdrawal. By maintaining individuals in their allocated intervention for analyses, the benefits of randomization will be captured ( 18 , 26 – 29 ). If analysis of actual treatment is solely relied upon, then some of the theoretical benefits of randomization may be lost. This analysis method relies on complete data. There are different approaches regarding the handling of missing data and no consensus has been put forth in the literature. Common approaches are imputation or carrying forward the last observed data from individuals to address issues of missing data ( 18 , 19 ).
Assessment timing can play an important role in the impact of interventions, particularly if intervention effects are acute and short lived ( 26 – 29 , 33 ). The specific timing of assessments are unique to each intervention, however, studies that allow for meaningfully different timing of assessments are subject to erroneous results. For example, if assessments occur differentially after an injection of a particularly fast acting, short-lived medication the difference observed between intervention arms may be due to a higher proportion of participants in one intervention arm being assessed hours after the intervention instead of minutes. By tracking differences in assessment times, researchers can address the potential scope of this problem, and try to address it using statistical or other methods ( 26 – 28 , 33 ).
Randomized controlled trials are the principle method for improving treatment of disease, and there are some standardized methods for grading RCTs, and subsequently creating best practice guidelines ( 29 , 34 – 36 ). Much of the current practice of medicine lacks moderate or high quality RCTs to address what treatment methods have demonstrated efficacy and much of the best practice guidelines remains based on consensus from experts ( 28 , 37 ). The reliance on high quality methodology in all types of studies will allow for continued improvement in the assessment of causal factors for health outcomes and the treatment of diseases.
There are many published standards for the design, execution and reporting of biomedical research, which can be found in Table 3 . The purpose and content of these standards and guidelines are to improve the quality of biomedical research which will result in providing sound conclusions to base medical decision making upon. There are published standards for categories of study designs such as observational studies (e.g. STROBE), interventional studies (e.g. CONSORT), diagnostic studies (e.g. STARD, QUADAS), systematic reviews and meta-analyses (e.g. PRISMA ), as well as others. The aim of these standards and guideline are to systematize and elevate the quality of biomedical research design, execution, and reporting.
Published standard for study design and reporting.
Consolidated Standards Of Reporting Trials | CONSORT | |
Strengthening the Reporting of Observational studies in Epidemiology | STROBE | |
Standards for Reporting Studies of Diagnostic Accuracy | STARD | |
Quality assessment of diagnostic accuracy studies | QUADAS | |
Preferred Reporting Items for Systematic Reviews and Meta-Analyses | PRISMA | |
Consolidated criteria for reporting qualitative research | COREQ | |
Statistical Analyses and Methods in the Published Literature | SAMPL | |
Consensus-based Clinical Case Reporting Guideline Development | CARE | |
Standards for Quality Improvement Reporting Excellence | SQUIRE | |
Consolidated Health Economic Evaluation Reporting Standards | CHEERS | |
Enhancing transparency in reporting the synthesis of qualitative research | ENTREQ |
When designing or evaluating a study it may be helpful to review the applicable standards prior to executing and publishing the study. All published standards and guidelines are available on the web, and are updated based on current best practices as biomedical research evolves. Additionally, there is a network called “Enhancing the quality and transparency of health research” (EQUATOR, www.equator-network.org ) , which has guidelines and checklists for all standards reported in Table 3 and is continually updated with new study design or specialty specific standards.
The appropriate selection of a study design is only one element in successful research. The selection of a study design should incorporate consideration of costs, access to cases, identification of the exposure, the epidemiologic measures that are required, and the level of evidence that is currently published regarding the specific exposure-outcome relationship that is being assessed. Reviewing appropriate published standards when designing a study can substantially strengthen the execution and interpretation of study results.
Potential conflict of interest
None declared.
Table of Contents
Most people think of a traditional experimental design when they consider research and published research papers. There is, however, a type of research that is more observational in nature, and it is appropriately referred to as “observational studies.”
There are many valuable reasons to utilize an observational study design. But, just as in research experimental design, different methods can be used when you’re considering this type of study. In this article, we’ll look at the advantages and disadvantages of an observational study design, as well as the 3 types of observational studies.
An observational study is when researchers are looking at the effect of some type of intervention, risk, a diagnostic test or treatment, without trying to manipulate who is, or who isn’t, exposed to it.
This differs from an experimental study, where the scientists are manipulating who is exposed to the treatment, intervention, etc., by having a control group, or those who are not exposed, and an experimental group, or those who are exposed to the intervention, treatment, etc. In the best studies, the groups are randomized, or chosen by chance.
Any evidence derived from systematic reviews is considered the best in the hierarchy of evidence, which considers which studies are deemed the most reliable. Next would be any evidence that comes from randomized controlled trials. Cohort studies and case studies follow, in that order.
Cohort studies and case studies are considered observational in design, whereas the randomized controlled trial would be an experimental study.
Let’s take a closer look at the different types of observational study design.
The different types of observational studies are used for different reasons. Selecting the best type for your research is critical to a successful outcome. One of the main reasons observational studies are used is when a randomized experiment would be considered unethical. For example, a life-saving medication used in a public health emergency. They are also used when looking at aetiology, or the cause of a condition or disease, as well as the treatment of rare conditions.
Researchers in case control studies identify individuals with an existing health issue or condition, or “cases,” along with a similar group without the condition, or “controls.” These two groups are then compared to identify predictors and outcomes. This type of study is helpful to generate a hypothesis that can then be researched.
This type of observational study is often used to help understand cause and effect. A cohort observational study looks at causes, incidence and prognosis, for example. A cohort is a group of people who are linked in a particular way, for example, a birth cohort would include people who were born within a specific period of time. Scientists might compare what happens to the members of the cohort who have been exposed to some variable to what occurs with members of the cohort who haven’t been exposed.
Unlike a cohort observational study, a cross sectional observational study does not explore cause and effect, but instead looks at prevalence. Here you would look at data from a particular group at one very specific period of time. Researchers would simply observe and record information about something present in the population, without manipulating any variables or interventions. These types of studies are commonly used in psychology, education and social science.
Observational study designs have the distinct advantage of allowing researchers to explore answers to questions where a randomized controlled trial, or RCT, would be unethical. Additionally, if the study is focused on a rare condition, studying existing cases as compared to non-affected individuals might be the most effective way to identify possible causes of the condition. Likewise, if very little is known about a condition or circumstance, a cohort study would be a good study design choice.
A primary advantage to the observational study design is that they can generally be completed quickly and inexpensively. A RCT can take years before the data is compiled and available. RCTs are more complex and involved, requiring many more logistics and details to iron out, whereas an observational study can be more easily designed and completed.
The main disadvantage of observational study designs is that they’re more open to dispute than an RCT. Of particular concern would be confounding biases. This is when a cohort might share other characteristics that affect the outcome versus the outcome stated in the study. An example would be that people who practice good sleeping habits have less heart disease. But, maybe those who practice effective sleeping habits also, in general, eat better and exercise more.
Need help with your research writing? With our Language Editing Plus service , we’ll help you improve the flow and writing of your paper, including UNLIMITED editing support. Use the simulator below to check the price for your manuscript, using the total number of words of the document.
You may also like.
Input your search keywords and press Enter.
What is the difference between an observational study and an experiment.
The key difference between observational studies and experimental designs is that a well-done observational study does not influence the responses of participants, while experiments do have some sort of treatment condition applied to at least some participants by random assignment .
Attrition refers to participants leaving a study. It always happens to some extent—for example, in randomized controlled trials for medical research.
Differential attrition occurs when attrition or dropout rates differ systematically between the intervention and the control group . As a result, the characteristics of the participants who drop out differ from the characteristics of those who stay in the study. Because of this, study results may be biased .
Action research is conducted in order to solve a particular issue immediately, while case studies are often conducted over a longer period of time and focus more on observing and analyzing a particular ongoing phenomenon.
Action research is focused on solving a problem or informing individual and community-based knowledge in a way that impacts teaching, learning, and other related processes. It is less focused on contributing theoretical input, instead producing actionable input.
Action research is particularly popular with educators as a form of systematic inquiry because it prioritizes reflection and bridges the gap between theory and practice. Educators are able to simultaneously investigate an issue as they solve it, and the method is very iterative and flexible.
A cycle of inquiry is another name for action research . It is usually visualized in a spiral shape following a series of steps, such as “planning → acting → observing → reflecting.”
To make quantitative observations , you need to use instruments that are capable of measuring the quantity you want to observe. For example, you might use a ruler to measure the length of an object or a thermometer to measure its temperature.
Criterion validity and construct validity are both types of measurement validity . In other words, they both show you how accurately a method measures something.
While construct validity is the degree to which a test or other measurement method measures what it claims to measure, criterion validity is the degree to which a test can predictively (in the future) or concurrently (in the present) measure something.
Construct validity is often considered the overarching type of measurement validity . You need to have face validity , content validity , and criterion validity in order to achieve construct validity.
Convergent validity and discriminant validity are both subtypes of construct validity . Together, they help you evaluate whether a test measures the concept it was designed to measure.
You need to assess both in order to demonstrate construct validity. Neither one alone is sufficient for establishing construct validity.
Content validity shows you how accurately a test or other measurement method taps into the various aspects of the specific construct you are researching.
In other words, it helps you answer the question: “does the test measure all aspects of the construct I want to measure?” If it does, then the test has high content validity.
The higher the content validity, the more accurate the measurement of the construct.
If the test fails to include parts of the construct, or irrelevant parts are included, the validity of the instrument is threatened, which brings your results into question.
Face validity and content validity are similar in that they both evaluate how suitable the content of a test is. The difference is that face validity is subjective, and assesses content at surface level.
When a test has strong face validity, anyone would agree that the test’s questions appear to measure what they are intended to measure.
For example, looking at a 4th grade math test consisting of problems in which students have to add and multiply, most people would agree that it has strong face validity (i.e., it looks like a math test).
On the other hand, content validity evaluates how well a test represents all the aspects of a topic. Assessing content validity is more systematic and relies on expert evaluation. of each question, analyzing whether each one covers the aspects that the test was designed to cover.
A 4th grade math test would have high content validity if it covered all the skills taught in that grade. Experts(in this case, math teachers), would have to evaluate the content validity by comparing the test to the learning objectives.
Snowball sampling is a non-probability sampling method . Unlike probability sampling (which involves some form of random selection ), the initial individuals selected to be studied are the ones who recruit new participants.
Because not every member of the target population has an equal chance of being recruited into the sample, selection in snowball sampling is non-random.
Snowball sampling is a non-probability sampling method , where there is not an equal chance for every member of the population to be included in the sample .
This means that you cannot use inferential statistics and make generalizations —often the goal of quantitative research . As such, a snowball sample is not representative of the target population and is usually a better fit for qualitative research .
Snowball sampling relies on the use of referrals. Here, the researcher recruits one or more initial participants, who then recruit the next ones.
Participants share similar characteristics and/or know each other. Because of this, not every member of the population has an equal chance of being included in the sample, giving rise to sampling bias .
Snowball sampling is best used in the following cases:
The reproducibility and replicability of a study can be ensured by writing a transparent, detailed method section and using clear, unambiguous language.
Reproducibility and replicability are related terms.
Stratified sampling and quota sampling both involve dividing the population into subgroups and selecting units from each subgroup. The purpose in both cases is to select a representative sample and/or to allow comparisons between subgroups.
The main difference is that in stratified sampling, you draw a random sample from each subgroup ( probability sampling ). In quota sampling you select a predetermined number or proportion of units, in a non-random manner ( non-probability sampling ).
Purposive and convenience sampling are both sampling methods that are typically used in qualitative data collection.
A convenience sample is drawn from a source that is conveniently accessible to the researcher. Convenience sampling does not distinguish characteristics among the participants. On the other hand, purposive sampling focuses on selecting participants possessing characteristics associated with the research study.
The findings of studies based on either convenience or purposive sampling can only be generalized to the (sub)population from which the sample is drawn, and not to the entire population.
Random sampling or probability sampling is based on random selection. This means that each unit has an equal chance (i.e., equal probability) of being included in the sample.
On the other hand, convenience sampling involves stopping people at random, which means that not everyone has an equal chance of being selected depending on the place, time, or day you are collecting your data.
Convenience sampling and quota sampling are both non-probability sampling methods. They both use non-random criteria like availability, geographical proximity, or expert knowledge to recruit study participants.
However, in convenience sampling, you continue to sample units or cases until you reach the required sample size.
In quota sampling, you first need to divide your population of interest into subgroups (strata) and estimate their proportions (quota) in the population. Then you can start your data collection, using convenience sampling to recruit participants, until the proportions in each subgroup coincide with the estimated proportions in the population.
A sampling frame is a list of every member in the entire population . It is important that the sampling frame is as complete as possible, so that your sample accurately reflects your population.
Stratified and cluster sampling may look similar, but bear in mind that groups created in cluster sampling are heterogeneous , so the individual characteristics in the cluster vary. In contrast, groups created in stratified sampling are homogeneous , as units share characteristics.
Relatedly, in cluster sampling you randomly select entire groups and include all units of each group in your sample. However, in stratified sampling, you select some units of all groups and include them in your sample. In this way, both methods can ensure that your sample is representative of the target population .
A systematic review is secondary research because it uses existing research. You don’t collect new data yourself.
An observational study is a great choice for you if your research question is based purely on observations. If there are ethical, logistical, or practical concerns that prevent you from conducting a traditional experiment , an observational study may be a good choice. In an observational study, there is no interference or manipulation of the research subjects, as well as no control or treatment groups .
It’s often best to ask a variety of people to review your measurements. You can ask experts, such as other researchers, or laypeople, such as potential participants, to judge the face validity of tests.
While experts have a deep understanding of research methods , the people you’re studying can provide you with valuable insights you may have missed otherwise.
Face validity is important because it’s a simple first step to measuring the overall validity of a test or technique. It’s a relatively intuitive, quick, and easy way to start checking whether a new measure seems useful at first glance.
Good face validity means that anyone who reviews your measure says that it seems to be measuring what it’s supposed to. With poor face validity, someone reviewing your measure may be left confused about what you’re measuring and why you’re using this method.
Face validity is about whether a test appears to measure what it’s supposed to measure. This type of validity is concerned with whether a measure seems relevant and appropriate for what it’s assessing only on the surface.
Statistical analyses are often applied to test validity with data from your measures. You test convergent validity and discriminant validity with correlations to see if results from your test are positively or negatively related to those of other established tests.
You can also use regression analyses to assess whether your measure is actually predictive of outcomes that you expect it to predict theoretically. A regression analysis that supports your expectations strengthens your claim of construct validity .
When designing or evaluating a measure, construct validity helps you ensure you’re actually measuring the construct you’re interested in. If you don’t have construct validity, you may inadvertently measure unrelated or distinct constructs and lose precision in your research.
Construct validity is often considered the overarching type of measurement validity , because it covers all of the other types. You need to have face validity , content validity , and criterion validity to achieve construct validity.
Construct validity is about how well a test measures the concept it was designed to evaluate. It’s one of four types of measurement validity , which includes construct validity, face validity , and criterion validity.
There are two subtypes of construct validity.
Naturalistic observation is a valuable tool because of its flexibility, external validity , and suitability for topics that can’t be studied in a lab setting.
The downsides of naturalistic observation include its lack of scientific control , ethical considerations , and potential for bias from observers and subjects.
Naturalistic observation is a qualitative research method where you record the behaviors of your research subjects in real world settings. You avoid interfering or influencing anything in a naturalistic observation.
You can think of naturalistic observation as “people watching” with a purpose.
A dependent variable is what changes as a result of the independent variable manipulation in experiments . It’s what you’re interested in measuring, and it “depends” on your independent variable.
In statistics, dependent variables are also called:
An independent variable is the variable you manipulate, control, or vary in an experimental study to explore its effects. It’s called “independent” because it’s not influenced by any other variables in the study.
Independent variables are also called:
As a rule of thumb, questions related to thoughts, beliefs, and feelings work well in focus groups. Take your time formulating strong questions, paying special attention to phrasing. Be careful to avoid leading questions , which can bias your responses.
Overall, your focus group questions should be:
A structured interview is a data collection method that relies on asking questions in a set order to collect data on a topic. They are often quantitative in nature. Structured interviews are best used when:
More flexible interview options include semi-structured interviews , unstructured interviews , and focus groups .
Social desirability bias is the tendency for interview participants to give responses that will be viewed favorably by the interviewer or other participants. It occurs in all types of interviews and surveys , but is most common in semi-structured interviews , unstructured interviews , and focus groups .
Social desirability bias can be mitigated by ensuring participants feel at ease and comfortable sharing their views. Make sure to pay attention to your own body language and any physical or verbal cues, such as nodding or widening your eyes.
This type of bias can also occur in observations if the participants know they’re being observed. They might alter their behavior accordingly.
The interviewer effect is a type of bias that emerges when a characteristic of an interviewer (race, age, gender identity, etc.) influences the responses given by the interviewee.
There is a risk of an interviewer effect in all types of interviews , but it can be mitigated by writing really high-quality interview questions.
A semi-structured interview is a blend of structured and unstructured types of interviews. Semi-structured interviews are best used when:
An unstructured interview is the most flexible type of interview, but it is not always the best fit for your research topic.
Unstructured interviews are best used when:
The four most common types of interviews are:
Deductive reasoning is commonly used in scientific research, and it’s especially associated with quantitative research .
In research, you might have come across something called the hypothetico-deductive method . It’s the scientific method of testing hypotheses to check whether your predictions are substantiated by real-world data.
Deductive reasoning is a logical approach where you progress from general ideas to specific conclusions. It’s often contrasted with inductive reasoning , where you start with specific observations and form general conclusions.
Deductive reasoning is also called deductive logic.
There are many different types of inductive reasoning that people use formally or informally.
Here are a few common types:
Inductive reasoning is a bottom-up approach, while deductive reasoning is top-down.
Inductive reasoning takes you from the specific to the general, while in deductive reasoning, you make inferences by going from general premises to specific conclusions.
In inductive research , you start by making observations or gathering data. Then, you take a broad scan of your data and search for patterns. Finally, you make general conclusions that you might incorporate into theories.
Inductive reasoning is a method of drawing conclusions by going from the specific to the general. It’s usually contrasted with deductive reasoning, where you proceed from general information to specific conclusions.
Inductive reasoning is also called inductive logic or bottom-up reasoning.
A hypothesis states your predictions about what your research will find. It is a tentative answer to your research question that has not yet been tested. For some research projects, you might have to write several hypotheses that address different aspects of your research question.
A hypothesis is not just a guess — it should be based on existing theories and knowledge. It also has to be testable, which means you can support or refute it through scientific research methods (such as experiments, observations and statistical analysis of data).
Triangulation can help:
But triangulation can also pose problems:
There are four main types of triangulation :
Many academic fields use peer review , largely to determine whether a manuscript is suitable for publication. Peer review enhances the credibility of the published manuscript.
However, peer review is also common in non-academic settings. The United Nations, the European Union, and many individual nations use peer review to evaluate grant applications. It is also widely used in medical and health-related fields as a teaching or quality-of-care measure.
Peer assessment is often used in the classroom as a pedagogical tool. Both receiving feedback and providing it are thought to enhance the learning process, helping students think critically and collaboratively.
Peer review can stop obviously problematic, falsified, or otherwise untrustworthy research from being published. It also represents an excellent opportunity to get feedback from renowned experts in your field. It acts as a first defense, helping you ensure your argument is clear and that there are no gaps, vague terms, or unanswered questions for readers who weren’t involved in the research process.
Peer-reviewed articles are considered a highly credible source due to this stringent process they go through before publication.
In general, the peer review process follows the following steps:
Exploratory research is often used when the issue you’re studying is new or when the data collection process is challenging for some reason.
You can use exploratory research if you have a general idea or a specific question that you want to study but there is no preexisting knowledge or paradigm with which to study it.
Exploratory research is a methodology approach that explores research questions that have not previously been studied in depth. It is often used when the issue you’re studying is new, or the data collection process is challenging in some way.
Explanatory research is used to investigate how or why a phenomenon occurs. Therefore, this type of research is often one of the first stages in the research process , serving as a jumping-off point for future research.
Exploratory research aims to explore the main aspects of an under-researched problem, while explanatory research aims to explain the causes and consequences of a well-defined problem.
Explanatory research is a research method used to investigate how or why something occurs when only a small amount of information is available pertaining to that topic. It can help you increase your understanding of a given topic.
Clean data are valid, accurate, complete, consistent, unique, and uniform. Dirty data include inconsistencies and errors.
Dirty data can come from any part of the research process, including poor research design , inappropriate measurement materials, or flawed data entry.
Data cleaning takes place between data collection and data analyses. But you can use some methods even before collecting data.
For clean data, you should start by designing measures that collect valid data. Data validation at the time of data entry or collection helps you minimize the amount of data cleaning you’ll need to do.
After data collection, you can use data standardization and data transformation to clean your data. You’ll also deal with any missing values, outliers, and duplicate values.
Every dataset requires different techniques to clean dirty data , but you need to address these issues in a systematic way. You focus on finding and resolving data points that don’t agree or fit with the rest of your dataset.
These data might be missing values, outliers, duplicate values, incorrectly formatted, or irrelevant. You’ll start with screening and diagnosing your data. Then, you’ll often standardize and accept or remove data to make your dataset consistent and valid.
Data cleaning is necessary for valid and appropriate analyses. Dirty data contain inconsistencies or errors , but cleaning your data helps you minimize or resolve these.
Without data cleaning, you could end up with a Type I or II error in your conclusion. These types of erroneous conclusions can be practically significant with important consequences, because they lead to misplaced investments or missed opportunities.
Data cleaning involves spotting and resolving potential data inconsistencies or errors to improve your data quality. An error is any value (e.g., recorded weight) that doesn’t reflect the true value (e.g., actual weight) of something that’s being measured.
In this process, you review, analyze, detect, modify, or remove “dirty” data to make your dataset “clean.” Data cleaning is also called data cleansing or data scrubbing.
Research misconduct means making up or falsifying data, manipulating data analyses, or misrepresenting results in research reports. It’s a form of academic fraud.
These actions are committed intentionally and can have serious consequences; research misconduct is not a simple mistake or a point of disagreement but a serious ethical failure.
Anonymity means you don’t know who the participants are, while confidentiality means you know who they are but remove identifying information from your research report. Both are important ethical considerations .
You can only guarantee anonymity by not collecting any personally identifying information—for example, names, phone numbers, email addresses, IP addresses, physical characteristics, photos, or videos.
You can keep data confidential by using aggregate information in your research report, so that you only refer to groups of participants rather than individuals.
Research ethics matter for scientific integrity, human rights and dignity, and collaboration between science and society. These principles make sure that participation in studies is voluntary, informed, and safe.
Ethical considerations in research are a set of principles that guide your research designs and practices. These principles include voluntary participation, informed consent, anonymity, confidentiality, potential for harm, and results communication.
Scientists and researchers must always adhere to a certain code of conduct when collecting data from others .
These considerations protect the rights of research participants, enhance research validity , and maintain scientific integrity.
In multistage sampling , you can use probability or non-probability sampling methods .
For a probability sample, you have to conduct probability sampling at every stage.
You can mix it up by using simple random sampling , systematic sampling , or stratified sampling to select units at different stages, depending on what is applicable and relevant to your study.
Multistage sampling can simplify data collection when you have large, geographically spread samples, and you can obtain a probability sample without a complete sampling frame.
But multistage sampling may not lead to a representative sample, and larger samples are needed for multistage samples to achieve the statistical properties of simple random samples .
These are four of the most common mixed methods designs :
Triangulation in research means using multiple datasets, methods, theories and/or investigators to address a research question. It’s a research strategy that can help you enhance the validity and credibility of your findings.
Triangulation is mainly used in qualitative research , but it’s also commonly applied in quantitative research . Mixed methods research always uses triangulation.
In multistage sampling , or multistage cluster sampling, you draw a sample from a population using smaller and smaller groups at each stage.
This method is often used to collect data from a large, geographically spread group of people in national surveys, for example. You take advantage of hierarchical groupings (e.g., from state to city to neighborhood) to create a sample that’s less expensive and time-consuming to collect data from.
No, the steepness or slope of the line isn’t related to the correlation coefficient value. The correlation coefficient only tells you how closely your data fit on a line, so two datasets with the same correlation coefficient can have very different slopes.
To find the slope of the line, you’ll need to perform a regression analysis .
Correlation coefficients always range between -1 and 1.
The sign of the coefficient tells you the direction of the relationship: a positive value means the variables change together in the same direction, while a negative value means they change together in opposite directions.
The absolute value of a number is equal to the number without its sign. The absolute value of a correlation coefficient tells you the magnitude of the correlation: the greater the absolute value, the stronger the correlation.
These are the assumptions your data must meet if you want to use Pearson’s r :
Quantitative research designs can be divided into two main categories:
Qualitative research designs tend to be more flexible. Common types of qualitative design include case study , ethnography , and grounded theory designs.
A well-planned research design helps ensure that your methods match your research aims, that you collect high-quality data, and that you use the right kind of analysis to answer your questions, utilizing credible sources . This allows you to draw valid , trustworthy conclusions.
The priorities of a research design can vary depending on the field, but you usually have to specify:
A research design is a strategy for answering your research question . It defines your overall approach and determines how you will collect and analyze data.
Questionnaires can be self-administered or researcher-administered.
Self-administered questionnaires can be delivered online or in paper-and-pen formats, in person or through mail. All questions are standardized so that all respondents receive the same questions with identical wording.
Researcher-administered questionnaires are interviews that take place by phone, in-person, or online between researchers and respondents. You can gain deeper insights by clarifying questions for respondents or asking follow-up questions.
You can organize the questions logically, with a clear progression from simple to complex, or randomly between respondents. A logical flow helps respondents process the questionnaire easier and quicker, but it may lead to bias. Randomization can minimize the bias from order effects.
Closed-ended, or restricted-choice, questions offer respondents a fixed set of choices to select from. These questions are easier to answer quickly.
Open-ended or long-form questions allow respondents to answer in their own words. Because there are no restrictions on their choices, respondents can answer in ways that researchers may not have otherwise considered.
A questionnaire is a data collection tool or instrument, while a survey is an overarching research method that involves collecting and analyzing data from people using questionnaires.
The third variable and directionality problems are two main reasons why correlation isn’t causation .
The third variable problem means that a confounding variable affects both variables to make them seem causally related when they are not.
The directionality problem is when two variables correlate and might actually have a causal relationship, but it’s impossible to conclude which variable causes changes in the other.
Correlation describes an association between variables : when one variable changes, so does the other. A correlation is a statistical indicator of the relationship between variables.
Causation means that changes in one variable brings about changes in the other (i.e., there is a cause-and-effect relationship between variables). The two variables are correlated with each other, and there’s also a causal link between them.
While causation and correlation can exist simultaneously, correlation does not imply causation. In other words, correlation is simply a relationship where A relates to B—but A doesn’t necessarily cause B to happen (or vice versa). Mistaking correlation for causation is a common error and can lead to false cause fallacy .
Controlled experiments establish causality, whereas correlational studies only show associations between variables.
In general, correlational research is high in external validity while experimental research is high in internal validity .
A correlation is usually tested for two variables at a time, but you can test correlations between three or more variables.
A correlation coefficient is a single number that describes the strength and direction of the relationship between your variables.
Different types of correlation coefficients might be appropriate for your data based on their levels of measurement and distributions . The Pearson product-moment correlation coefficient (Pearson’s r ) is commonly used to assess a linear relationship between two quantitative variables.
A correlational research design investigates relationships between two variables (or more) without the researcher controlling or manipulating any of them. It’s a non-experimental type of quantitative research .
A correlation reflects the strength and/or direction of the association between two or more variables.
Random error is almost always present in scientific studies, even in highly controlled settings. While you can’t eradicate it completely, you can reduce random error by taking repeated measurements, using a large sample, and controlling extraneous variables .
You can avoid systematic error through careful design of your sampling , data collection , and analysis procedures. For example, use triangulation to measure your variables using multiple methods; regularly calibrate instruments or procedures; use random sampling and random assignment ; and apply masking (blinding) where possible.
Systematic error is generally a bigger problem in research.
With random error, multiple measurements will tend to cluster around the true value. When you’re collecting data from a large sample , the errors in different directions will cancel each other out.
Systematic errors are much more problematic because they can skew your data away from the true value. This can lead you to false conclusions ( Type I and II errors ) about the relationship between the variables you’re studying.
Random and systematic error are two types of measurement error.
Random error is a chance difference between the observed and true values of something (e.g., a researcher misreading a weighing scale records an incorrect measurement).
Systematic error is a consistent or proportional difference between the observed and true values of something (e.g., a miscalibrated scale consistently records weights as higher than they actually are).
On graphs, the explanatory variable is conventionally placed on the x-axis, while the response variable is placed on the y-axis.
The term “ explanatory variable ” is sometimes preferred over “ independent variable ” because, in real world contexts, independent variables are often influenced by other variables. This means they aren’t totally independent.
Multiple independent variables may also be correlated with each other, so “explanatory variables” is a more appropriate term.
The difference between explanatory and response variables is simple:
In a controlled experiment , all extraneous variables are held constant so that they can’t influence the results. Controlled experiments require:
Depending on your study topic, there are various other methods of controlling variables .
There are 4 main types of extraneous variables :
An extraneous variable is any variable that you’re not investigating that can potentially affect the dependent variable of your research study.
A confounding variable is a type of extraneous variable that not only affects the dependent variable, but is also related to the independent variable.
In a factorial design, multiple independent variables are tested.
If you test two variables, each level of one independent variable is combined with each level of the other independent variable to create different conditions.
Within-subjects designs have many potential threats to internal validity , but they are also very statistically powerful .
Advantages:
Disadvantages:
While a between-subjects design has fewer threats to internal validity , it also requires more participants for high statistical power than a within-subjects design .
Yes. Between-subjects and within-subjects designs can be combined in a single study when you have two or more independent variables (a factorial design). In a mixed factorial design, one variable is altered between subjects and another is altered within subjects.
In a between-subjects design , every participant experiences only one condition, and researchers assess group differences between participants in various conditions.
In a within-subjects design , each participant experiences all conditions, and researchers test the same participants repeatedly for differences between conditions.
The word “between” means that you’re comparing different conditions between groups, while the word “within” means you’re comparing different conditions within the same group.
Random assignment is used in experiments with a between-groups or independent measures design. In this research design, there’s usually a control group and one or more experimental groups. Random assignment helps ensure that the groups are comparable.
In general, you should always use random assignment in this type of experimental design when it is ethically possible and makes sense for your study topic.
To implement random assignment , assign a unique number to every member of your study’s sample .
Then, you can use a random number generator or a lottery method to randomly assign each number to a control or experimental group. You can also do so manually, by flipping a coin or rolling a dice to randomly assign participants to groups.
Random selection, or random sampling , is a way of selecting members of a population for your study’s sample.
In contrast, random assignment is a way of sorting the sample into control and experimental groups.
Random sampling enhances the external validity or generalizability of your results, while random assignment improves the internal validity of your study.
In experimental research, random assignment is a way of placing participants from your sample into different groups using randomization. With this method, every member of the sample has a known or equal chance of being placed in a control group or an experimental group.
“Controlling for a variable” means measuring extraneous variables and accounting for them statistically to remove their effects on other variables.
Researchers often model control variable data along with independent and dependent variable data in regression analyses and ANCOVAs . That way, you can isolate the control variable’s effects from the relationship between the variables of interest.
Control variables help you establish a correlational or causal relationship between variables by enhancing internal validity .
If you don’t control relevant extraneous variables , they may influence the outcomes of your study, and you may not be able to demonstrate that your results are really an effect of your independent variable .
A control variable is any variable that’s held constant in a research study. It’s not a variable of interest in the study, but it’s controlled because it could influence the outcomes.
Including mediators and moderators in your research helps you go beyond studying a simple relationship between two variables for a fuller picture of the real world. They are important to consider when studying complex correlational or causal relationships.
Mediators are part of the causal pathway of an effect, and they tell you how or why an effect takes place. Moderators usually help you judge the external validity of your study by identifying the limitations of when the relationship between variables holds.
If something is a mediating variable :
A confounder is a third variable that affects variables of interest and makes them seem related when they are not. In contrast, a mediator is the mechanism of a relationship between two variables: it explains the process by which they are related.
A mediator variable explains the process through which two variables are related, while a moderator variable affects the strength and direction of that relationship.
There are three key steps in systematic sampling :
Systematic sampling is a probability sampling method where researchers select members of the population at a regular interval – for example, by selecting every 15th person on a list of the population. If the population is in a random order, this can imitate the benefits of simple random sampling .
Yes, you can create a stratified sample using multiple characteristics, but you must ensure that every participant in your study belongs to one and only one subgroup. In this case, you multiply the numbers of subgroups for each characteristic to get the total number of groups.
For example, if you were stratifying by location with three subgroups (urban, rural, or suburban) and marital status with five subgroups (single, divorced, widowed, married, or partnered), you would have 3 x 5 = 15 subgroups.
You should use stratified sampling when your sample can be divided into mutually exclusive and exhaustive subgroups that you believe will take on different mean values for the variable that you’re studying.
Using stratified sampling will allow you to obtain more precise (with lower variance ) statistical estimates of whatever you are trying to measure.
For example, say you want to investigate how income differs based on educational attainment, but you know that this relationship can vary based on race. Using stratified sampling, you can ensure you obtain a large enough sample from each racial group, allowing you to draw more precise conclusions.
In stratified sampling , researchers divide subjects into subgroups called strata based on characteristics that they share (e.g., race, gender, educational attainment).
Once divided, each subgroup is randomly sampled using another probability sampling method.
Cluster sampling is more time- and cost-efficient than other probability sampling methods , particularly when it comes to large samples spread across a wide geographical area.
However, it provides less statistical certainty than other methods, such as simple random sampling , because it is difficult to ensure that your clusters properly represent the population as a whole.
There are three types of cluster sampling : single-stage, double-stage and multi-stage clustering. In all three types, you first divide the population into clusters, then randomly select clusters for use in your sample.
Cluster sampling is a probability sampling method in which you divide a population into clusters, such as districts or schools, and then randomly select some of these clusters as your sample.
The clusters should ideally each be mini-representations of the population as a whole.
If properly implemented, simple random sampling is usually the best sampling method for ensuring both internal and external validity . However, it can sometimes be impractical and expensive to implement, depending on the size of the population to be studied,
If you have a list of every member of the population and the ability to reach whichever members are selected, you can use simple random sampling.
The American Community Survey is an example of simple random sampling . In order to collect detailed data on the population of the US, the Census Bureau officials randomly select 3.5 million households per year and use a variety of methods to convince them to fill out the survey.
Simple random sampling is a type of probability sampling in which the researcher randomly selects a subset of participants from a population . Each member of the population has an equal chance of being selected. Data is then collected from as large a percentage as possible of this random subset.
Quasi-experimental design is most useful in situations where it would be unethical or impractical to run a true experiment .
Quasi-experiments have lower internal validity than true experiments, but they often have higher external validity as they can use real-world interventions instead of artificial laboratory settings.
A quasi-experiment is a type of research design that attempts to establish a cause-and-effect relationship. The main difference with a true experiment is that the groups are not randomly assigned.
Blinding is important to reduce research bias (e.g., observer bias , demand characteristics ) and ensure a study’s internal validity .
If participants know whether they are in a control or treatment group , they may adjust their behavior in ways that affect the outcome that researchers are trying to measure. If the people administering the treatment are aware of group assignment, they may treat participants differently and thus directly or indirectly influence the final results.
Blinding means hiding who is assigned to the treatment group and who is assigned to the control group in an experiment .
A true experiment (a.k.a. a controlled experiment) always includes at least one control group that doesn’t receive the experimental treatment.
However, some experiments use a within-subjects design to test treatments without a control group. In these designs, you usually compare one group’s outcomes before and after a treatment (instead of comparing outcomes between different groups).
For strong internal validity , it’s usually best to include a control group if possible. Without a control group, it’s harder to be certain that the outcome was caused by the experimental treatment and not by other variables.
An experimental group, also known as a treatment group, receives the treatment whose effect researchers wish to study, whereas a control group does not. They should be identical in all other ways.
Individual Likert-type questions are generally considered ordinal data , because the items have clear rank order, but don’t have an even distribution.
Overall Likert scale scores are sometimes treated as interval data. These scores are considered to have directionality and even spacing between them.
The type of data determines what statistical tests you should use to analyze your data.
A Likert scale is a rating scale that quantitatively assesses opinions, attitudes, or behaviors. It is made up of 4 or more questions that measure a single attitude or trait when response scores are combined.
To use a Likert scale in a survey , you present participants with Likert-type questions or statements, and a continuum of items, usually with 5 or 7 possible responses, to capture their degree of agreement.
In scientific research, concepts are the abstract ideas or phenomena that are being studied (e.g., educational achievement). Variables are properties or characteristics of the concept (e.g., performance at school), while indicators are ways of measuring or quantifying variables (e.g., yearly grade reports).
The process of turning abstract concepts into measurable variables and indicators is called operationalization .
There are various approaches to qualitative data analysis , but they all share five steps in common:
The specifics of each step depend on the focus of the analysis. Some common approaches include textual analysis , thematic analysis , and discourse analysis .
There are five common approaches to qualitative research :
Hypothesis testing is a formal procedure for investigating our ideas about the world using statistics. It is used by scientists to test specific predictions, called hypotheses , by calculating how likely it is that a pattern or relationship between variables could have arisen by chance.
Operationalization means turning abstract conceptual ideas into measurable observations.
For example, the concept of social anxiety isn’t directly observable, but it can be operationally defined in terms of self-rating scores, behavioral avoidance of crowded places, or physical anxiety symptoms in social situations.
Before collecting data , it’s important to consider how you will operationalize the variables that you want to measure.
When conducting research, collecting original data has significant advantages:
However, there are also some drawbacks: data collection can be time-consuming, labor-intensive and expensive. In some cases, it’s more efficient to use secondary data that has already been collected by someone else, but the data might be less reliable.
Data collection is the systematic process by which observations or measurements are gathered in research. It is used in many different contexts by academics, governments, businesses, and other organizations.
There are several methods you can use to decrease the impact of confounding variables on your research: restriction, matching, statistical control and randomization.
In restriction , you restrict your sample by only including certain subjects that have the same values of potential confounding variables.
In matching , you match each of the subjects in your treatment group with a counterpart in the comparison group. The matched subjects have the same values on any potential confounding variables, and only differ in the independent variable .
In statistical control , you include potential confounders as variables in your regression .
In randomization , you randomly assign the treatment (or independent variable) in your study to a sufficiently large number of subjects, which allows you to control for all potential confounding variables.
A confounding variable is closely related to both the independent and dependent variables in a study. An independent variable represents the supposed cause , while the dependent variable is the supposed effect . A confounding variable is a third variable that influences both the independent and dependent variables.
Failing to account for confounding variables can cause you to wrongly estimate the relationship between your independent and dependent variables.
To ensure the internal validity of your research, you must consider the impact of confounding variables. If you fail to account for them, you might over- or underestimate the causal relationship between your independent and dependent variables , or even find a causal relationship where none exists.
Yes, but including more than one of either type requires multiple research questions .
For example, if you are interested in the effect of a diet on health, you can use multiple measures of health: blood sugar, blood pressure, weight, pulse, and many more. Each of these is its own dependent variable with its own research question.
You could also choose to look at the effect of exercise levels as well as diet, or even the additional effect of the two combined. Each of these is a separate independent variable .
To ensure the internal validity of an experiment , you should only change one independent variable at a time.
No. The value of a dependent variable depends on an independent variable, so a variable cannot be both independent and dependent at the same time. It must be either the cause or the effect, not both!
You want to find out how blood sugar levels are affected by drinking diet soda and regular soda, so you conduct an experiment .
Determining cause and effect is one of the most important parts of scientific research. It’s essential to know which is the cause – the independent variable – and which is the effect – the dependent variable.
In non-probability sampling , the sample is selected based on non-random criteria, and not every member of the population has a chance of being included.
Common non-probability sampling methods include convenience sampling , voluntary response sampling, purposive sampling , snowball sampling, and quota sampling .
Probability sampling means that every member of the target population has a known chance of being included in the sample.
Probability sampling methods include simple random sampling , systematic sampling , stratified sampling , and cluster sampling .
Using careful research design and sampling procedures can help you avoid sampling bias . Oversampling can be used to correct undercoverage bias .
Some common types of sampling bias include self-selection bias , nonresponse bias , undercoverage bias , survivorship bias , pre-screening or advertising bias, and healthy user bias.
Sampling bias is a threat to external validity – it limits the generalizability of your findings to a broader group of people.
A sampling error is the difference between a population parameter and a sample statistic .
A statistic refers to measures about the sample , while a parameter refers to measures about the population .
Populations are used when a research question requires data from every member of the population. This is usually only feasible when the population is small and easily accessible.
Samples are used to make inferences about populations . Samples are easier to collect data from because they are practical, cost-effective, convenient, and manageable.
There are seven threats to external validity : selection bias , history, experimenter effect, Hawthorne effect , testing effect, aptitude-treatment and situation effect.
The two types of external validity are population validity (whether you can generalize to other groups of people) and ecological validity (whether you can generalize to other situations and settings).
The external validity of a study is the extent to which you can generalize your findings to different groups of people, situations, and measures.
Cross-sectional studies cannot establish a cause-and-effect relationship or analyze behavior over a period of time. To investigate cause and effect, you need to do a longitudinal study or an experimental study .
Cross-sectional studies are less expensive and time-consuming than many other types of study. They can provide useful insights into a population’s characteristics and identify correlations for further research.
Sometimes only cross-sectional data is available for analysis; other times your research question may only require a cross-sectional study to answer it.
Longitudinal studies can last anywhere from weeks to decades, although they tend to be at least a year long.
The 1970 British Cohort Study , which has collected data on the lives of 17,000 Brits since their births in 1970, is one well-known example of a longitudinal study .
Longitudinal studies are better to establish the correct sequence of events, identify changes over time, and provide insight into cause-and-effect relationships, but they also tend to be more expensive and time-consuming than other types of studies.
Longitudinal studies and cross-sectional studies are two different types of research design . In a cross-sectional study you collect data from a population at a specific point in time; in a longitudinal study you repeatedly collect data from the same sample over an extended period of time.
Longitudinal study | Cross-sectional study |
---|---|
observations | Observations at a in time |
Observes the multiple times | Observes (a “cross-section”) in the population |
Follows in participants over time | Provides of society at a given point |
There are eight threats to internal validity : history, maturation, instrumentation, testing, selection bias , regression to the mean, social interaction and attrition .
Internal validity is the extent to which you can be confident that a cause-and-effect relationship established in a study cannot be explained by other factors.
In mixed methods research , you use both qualitative and quantitative data collection and analysis methods to answer your research question .
The research methods you use depend on the type of data you need to answer your research question .
A confounding variable , also called a confounder or confounding factor, is a third variable in a study examining a potential cause-and-effect relationship.
A confounding variable is related to both the supposed cause and the supposed effect of the study. It can be difficult to separate the true effect of the independent variable from the effect of the confounding variable.
In your research design , it’s important to identify potential confounding variables and plan how you will reduce their impact.
Discrete and continuous variables are two types of quantitative variables :
Quantitative variables are any variables where the data represent amounts (e.g. height, weight, or age).
Categorical variables are any variables where the data represent groups. This includes rankings (e.g. finishing places in a race), classifications (e.g. brands of cereal), and binary outcomes (e.g. coin flips).
You need to know what type of variables you are working with to choose the right statistical test for your data and interpret your results .
You can think of independent and dependent variables in terms of cause and effect: an independent variable is the variable you think is the cause , while a dependent variable is the effect .
In an experiment, you manipulate the independent variable and measure the outcome in the dependent variable. For example, in an experiment about the effect of nutrients on crop growth:
Defining your variables, and deciding how you will manipulate and measure them, is an important part of experimental design .
Experimental design means planning a set of procedures to investigate a relationship between variables . To design a controlled experiment, you need:
When designing the experiment, you decide:
Experimental design is essential to the internal and external validity of your experiment.
I nternal validity is the degree of confidence that the causal relationship you are testing is not influenced by other factors or variables .
External validity is the extent to which your results can be generalized to other contexts.
The validity of your experiment depends on your experimental design .
Reliability and validity are both about how well a method measures something:
If you are doing experimental research, you also have to consider the internal and external validity of your experiment.
A sample is a subset of individuals from a larger population . Sampling means selecting the group that you will actually collect data from in your research. For example, if you are researching the opinions of students in your university, you could survey a sample of 100 students.
In statistics, sampling allows you to test a hypothesis about the characteristics of a population.
Quantitative research deals with numbers and statistics, while qualitative research deals with words and meanings.
Quantitative methods allow you to systematically measure variables and test hypotheses . Qualitative methods allow you to explore concepts and experiences in more detail.
Methodology refers to the overarching strategy and rationale of your research project . It involves studying the methods used in your field and the theories or principles behind them, in order to develop an approach that matches your objectives.
Methods are the specific tools and procedures you use to collect and analyze data (for example, experiments, surveys , and statistical tests ).
In shorter scientific papers, where the aim is to report the findings of a specific study, you might simply describe what you did in a methods section .
In a longer or more complex research project, such as a thesis or dissertation , you will probably include a methodology section , where you explain your approach to answering the research questions and cite relevant sources to support your choice of methods.
Want to contact us directly? No problem. We are always here for you.
Our team helps students graduate by offering:
Scribbr specializes in editing study-related documents . We proofread:
Scribbr’s Plagiarism Checker is powered by elements of Turnitin’s Similarity Checker , namely the plagiarism detection software and the Internet Archive and Premium Scholarly Publications content databases .
The add-on AI detector is powered by Scribbr’s proprietary software.
The Scribbr Citation Generator is developed using the open-source Citation Style Language (CSL) project and Frank Bennett’s citeproc-js . It’s the same technology used by dozens of other popular citation tools, including Mendeley and Zotero.
You can find all the citation styles and locales used in the Scribbr Citation Generator in our publicly accessible repository on Github .
When we read about research studies and reports, many are times that we fail to pay attention to the design of the study. For you to know the quality of the research findings, it is paramount to start by understanding some basics of research/study design.
The primary goal of doing a study is to evaluate the relationship between several variables. For example, does eating fast food result in teenagers being overweight? Or does going to college increase the chances of getting a job? Most studies fall into two main categories, observational and experimental studies, but what is the difference? Other widely accepted research types are cohort studies, randomized controls, and case-control studies, but these three are part of either experimental or observational study. Keep reading to understand the difference between observational study and experiment.
To understand observational study vs experiment, let us start by looking at each of them.
So, what is an observational study ? This is a form of research where the measurement is done on the selected sample without running a control experiment. Therefore, the researcher observes the impact of a specific risk factor, such as treatment or intervention, without focusing on who is not exposed. It is simply a matter of observing what is happening.
When an observational report is released, it indicates that there might be a relationship between several variables, but this cannot be relied on. It is simply too weak or biased. We will demonstrate this with an example.
A study asking people how they liked a new film that was released a few months ago is a good example of an observational study. The researcher in the study does not have any control over the participants. Therefore, even if the study point to some relationship between the main variables, it is considered too weak. For example, the study did not factor in the possibility of viewers watching other films.
The main difference between an observational study and an experiment is that the latter is randomized . Again, unlike the observational study statistics, which are considered biased and weak, evidence from experimental research is stronger.
If you are thinking of carrying a research and have been wondering whether to go for randomized experiment vs observational study, here are some key advantages of the latter.
While the advantages of observational research might appear attractive, you need to weigh them against the cons. To run conclusive observational research, you might require a lot of time. Sometimes, this might run for years or decades.
The results from observational studies are also open to a lot of criticism because of confounding biases. For example, a cohort study might conclude that most people who love to meditate regularly suffer less from heart issues. However, this alone might not be the only cause of low cases of heart problems. The people who medicate might also be following healthy diets and doing a lot of exercises to stay healthy.
Observational studies further branches into several categories, including cohort study, cross-sectional, and case-control. Here is a breakdown of these different types of studies:
For study purposes, a “cohort” is a team or group of people who are somehow linked. Example, people born within a specific period might be referred to as a “birth cohort.”
The concept of cohort study edges close to that of experimental research. Here, the researcher records whether every participant in the cohort is affected by the selected variables. In a medical setting, the researcher might want to know whether the cohort population in the study got exposed to a certain variable and if they developed the medical condition of interest. This is the most preferred method of study when urgent response, especially to a public health concern, such as a disease outbreak is reported.
It is important to appreciate that this is different from experimental research because the investigator simply observes but does not determine the exposure status of the participants.
In this type of study, the researcher enrolls people with a health issue and another without the problem. Then, the two groups are compared based on exposure. The control group is used to generate an estimate of the expected exposure in the population.
This is the third type of observational type of study, and it involves taking a sample from a population that is exposed to health risk and measuring them to establish the extent of the outcome. This study is very common in health settings when researchers want to know the prevalence of a health condition at any specific moment. For example, in a cross-sectional study, some of the selected persons might have lived with high blood pressure for years, while others might have started seeing the signs recently.
Now that you know the observational study definition, we will now compare it with experiment research. So, what is experimental research?
In experimental design, the researcher randomly assigns a selected part of the population some treatment to make a cause and effect conclusion. The random selection of samples is largely what makes the experiment different from the observational study design.
The researcher controls the environment, such as exposure levels, and then checks the response produced by the population. In science, the evidence generated by experimental studies is stronger and less contested compared to that produced by observational studies.
Sometimes, you might find experimental study design being referred to as a scientific study. Always remember that when using experimental studies, you need two groups, the main experiment group (part of the population exposed to a variable) and the control (another group that does not get exposed/ treatment by the researcher).
Here are the main advantages to expect for using experimental study vs observational experiment.
When using experimental studies, it is important to appreciate that it can be pretty expensive because you are essentially following two groups, the experiment sample and control. The cost also arises from the factor that you might need to control the exposure levels and closely follow the progress before drawing a conclusion.
Now that we have looked at how each design, experimental and observational, work, we will now turn to examples and identify their application.
To improve the quality of life, many people are trying to quit smoking by following different strategies, but it is true that quitting is not easy. So the methods that are used by smokers include:
The variable in the study is (I, II, III, IV), and the outcome or response is success or failure to quit the problem of smoking. If you select to use an observational method, the values of the variables (I, ii, iii, iv) would happen naturally, meaning that you would not control them. In an experimental study, values would be assigned by the researcher, implying that you would tell the participants the methods to use. Here is a demonstration:
The results from the experimental study might be as shown below:
Quit smoking successfully | Failed to quit smoking | Total number of participants | Percentage of those who quit smoking | |
Drug and therapy | 83 | 167 | 250 | 33% |
Drugs only | 60 | 190 | 250 | 24% |
Therapy only | 59 | 191 | 250 | 24% |
Cold turkey | 12 | 238 | 250 | 5% |
From the results of the experimental study, we can say that combining therapy and drugs method helped most smokers to quit the habit successfully. Therefore, a policy can be developed to adopt the most successful method for helping smokers quit the problem.
It is important to note that both studies commence with a random sample. The difference between an observational study and an experiment is that the sample is divided in the latter while it is not in the former. In the case of the experimental study, the researcher is controlling the main variables and then checking the relationship.
A researcher picked a random sample of learners in a class and asked them about their study habits at home. The data showed that students who used at least 30 minutes to study after school scored better grades than those who never studied at all.
This type of study can be classified as observational because the researcher simply asked the respondents about their study habits after school. Because there was no group given a particular treatment, the study cannot qualify as experimental.
In another study, the researcher randomly picked two groups of students in school to determine the effectiveness of a new study method. Group one was asked to follow the new method for a period of three months, while the other was asked to simply study the way they were used. Then, the researcher checked the scores between the two groups to determine if the new method is better.
So, is this an experimental or observational study? This type of study can be categorized as experimental because the researcher randomly picked two groups of respondents. Then, one group was given some treatment, and the other one was not.
In one of the studies, the researcher took a random sample of people and looked at their eating habits. Then, every member was classified as either healthy or at risk of developing obesity. The researcher also drew recommendations to help people at risk of developing overweight issues to avoid the problem.
This type of study is observational because the researcher took a random sample but did no accord any group a special treatment. The study simply observed the people’s eating habits and classified them.
In one of the studies done in Japan, the researcher wanted to know the levels of radioactive materials in people’s tissues after the bombing of Hiroshima and Nagasaki in 1945. Therefore, he took a random sample of 1000 people in the region and asked them to get checked to determine the levels of radiation in their tissues.
After the study, the researcher concluded that the level of radiation in people’s tissues is still very high and might be associated with different types of diseases being reported in the region. Can you determine what type of study design this is?
The research is an example observational study because it did not have any control. The researcher only observed the levels but did not have any type of control group. Again, there was no special treatment to one of the study populations.
If you are a researcher, it is very important to be able to define observational study and experiment research before commencing your work. This can help you to determine the different parameters and how to go about the study. As we have demonstrated, observational studies mainly involve gathering the findings from the field without trying to control the variables. Although this study’s results can be contested, it is the most recommended method when using other studies such as experimental design, is unfeasible or unethical.
Experimental studies giving the researcher greater control over the study population by controlling the variables. Although more expensive, it takes a relatively shorter time, and results are less biased.
Now, go ahead and design your study. Always remember that you can seek help from either your lecturer or an expert when designing the study. Once you understand the concept of observational study vs experiment well, research can become so enjoyable and fun.
Your email address will not be published. Required fields are marked *
Comment * Error message
Name * Error message
Email * Error message
Save my name, email, and website in this browser for the next time I comment.
As Putin continues killing civilians, bombing kindergartens, and threatening WWIII, Ukraine fights for the world's peaceful future.
Ukraine Live Updates
In order to continue enjoying our site, we ask that you confirm your identity as a human. Thank you very much for your cooperation.
The stripping of carbon nanotube forests (CNTF) is an urgent problem in terms of its application in thermal management and semiconductor devices. The easy‑peel and vertically aligned CNTF is prepared in a two‑step process of high vacuum magnetron sputtering and chemical vapor deposition. Based on the thick alumina buffer layer, CNTF with heights of 100 μm, 300 μm, and 500 μm was prepared by adjusting the magnetron sputtering power. Through SEM observation and calculation, the corresponding densities were 2.5 × 109 cm−2, 6.4 × 109 cm−2, and 1.21 × 1010 cm−2, with an average curvature of 1.64 × 102 m−1, 1.35 × 102 m−1 and 0.53 × 102 m‑−1, respectively. The TEM, XPS, Raman, and TGA were used to investigate the growth mechanism of CNTF. The experimental results show that the catalyst particles annealed under high sputtering power refined with high density can grow a low‑defect, well‑oriented, and high‑density CNTF, and confirm the tip growth route. Further analysis shows that the easy‑peel properties of CNTF depend on the thickness of the alumina layer, the tip growth route, and the tight entanglement between the nanotubes. This paper provides technical guidance and support for the preparation, stripping, and application of monolithic CNTF.
Respiratory Research volume 25 , Article number: 286 ( 2024 ) Cite this article
126 Accesses
Metrics details
The use of machine learning(ML) methods would improve the diagnosis of small airway dysfunction(SAD) in subjects with chronic respiratory symptoms and preserved pulmonary function(PPF). This paper evaluated the performance of several ML algorithms associated with the impulse oscillometry(IOS) analysis to aid in the diagnostic of respiratory changes in SAD. We also find out the best configuration for this task.
IOS and spirometry were measured in 280 subjects, including a healthy control group ( n = 78), a group with normal spirometry ( n = 158) and a group with abnormal spirometry ( n = 44). Various supervised machine learning (ML) algorithms and feature selection strategies were examined, such as Support Vector Machines (SVM), Random Forests (RF), Adaptive Boosting (ADABOOST), Navie Bayesian (BAYES), and K-Nearest Neighbors (KNN).
The first experiment of this study demonstrated that the best oscillometric parameter (BOP) was R5, with an AUC value of 0.642, when comparing a healthy control group(CG) with patients in the group without lung volume-defined SAD(PPFN). The AUC value of BOP in the control group was 0.769 compared with patients with spirometry defined SAD(PPFA) in the PPF population. In the second experiment, the ML technique was used. In CGvsPPFN, RF and ADABOOST had the best diagnostic results (AUC = 0.914, 0.915), with significantly higher accuracy compared to BOP ( p < 0.01). In CGvsPPFA, RF and ADABOOST had the best diagnostic results (AUC = 0.951, 0.971) and significantly higher diagnostic accuracy ( p < 0.01). In the third, fourth and fifth experiments, different feature selection techniques allowed us to find the best IOS parameters (R5, (R5-R20)/R5 and Fres). The results demonstrate that the performance of ADABOOST remained essentially unaltered following the application of the feature selector, whereas the diagnostic accuracy of the remaining four classifiers (RF, SVM, BAYES, and KNN) is marginally enhanced.
IOS combined with ML algorithms provide a new method for diagnosing SAD in subjects with chronic respiratory symptoms and PPF. The present study’s findings provide evidence that this combination may help in the early diagnosis of respiratory changes in these patients.
Two major chronic respiratory disorders that can affect the small airways include asthma and chronic obstructive pulmonary disease (COPD). Evidence from prospective studies indicates that asthma and COPD may occur before small airway dysfunction (SAD) [ 1 , 2 , 3 ]. Symptoms of COPD and asthma include coughing, producing phlegm, dyspnea, and wheezing. The following symptoms may indicate SAD in some subjects: negative airway hyperresponsiveness (AHR) or bronchial reversibility (BR), which means the subject does not meet the pulmonary function criteria for COPD or asthma, and preserved pulmonary function (PPF, forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) ratio ≥ 0.70 [ 4 ]). According to a large-scale multi-stage stratified sampling survey, about 40% of Chinese individuals 20 years of age and older have spirometrically characterized SAD [ 5 ]. Owing to the severe impact of SAD, it was crucial to identify and treat the condition early.
The “quiet zone” is made up of small airways (with an inner diameter of less than 2 mm), which had a huge cross-sectional area and contribute very little to the total airway resistance. [ 6 ] In clinical practice, spirometry was the most widely used technique to assess small-airway function. The parameters that were employed include FVC50% (FEF50%), FVC75% at expiration (FEF75%), and forced expiratory flow between 25% and 75% of FVC (FEF25–75%). At least two of the three small airway markers (FEF25–75%, FEF50%, and FEF75%) had a projected value of less than 65%, which was the definition of spirometry SAD [ 5 ]. However, spirometry requires good cooperation of subjects, and the great variability of values makes its reliability not universally accepted [ 7 , 8 ]. An approach to measuring respiratory impedance based on the forced oscillation technique (FOT) is called impulse oscillometry (IOS). All that is needed for the IOS measurement is quiet tidal breathing, which is easy to do, appropriate for a broad spectrum of individuals, and yields a variety of respiratory physiological data. IOS is able to measure the respiratory mechanics during quiet tidal breathing, which sets it apart from spirometry. Because of externally overlaid oscillatory signals, it is independent of subject effort, unlike spirometry [ 9 ]. Furthermore, it appears to correlate better with small airway features and may be more sensitive in identifying SAD [ 10 , 11 , 12 ]. Since IOS can reflect the viscosity of the respiratory system through electrical resistance (RRS) and the elastic and inertial properties of the respiratory system through reactance (XRS), it can be combined with spirometry to gain more insight into individual pathological changes.
IOS was not currently frequently utilized in pulmonary function assessment, though. This approach’s drawback stems from the fact that it relied on electrical engineering ideas, which might be challenging to interpret in a clinical context. Another important consideration is the expensive inspection apparatus. Therefore, even though the IOS test is straightforward, a busy, inexperienced pulmonary function technician or primary care physician would find it challenging to interpret the resistance and reactance curves, as well as the derived values, without proper training and expertise. Furthermore, the analysis is challenging due to the findings for the IOS test values being dispersed. Consequently, machine learning (ML)-based computer-aided decision systems can enhance the functionality of IOS and support physicians in strengthening the diagnosis, monitoring, and treatment of chronic respiratory disorders, such as asthma and COPD.
In this context, we hypothesized that the use of ML methods in combination with IOS test would improve the diagnosis of small airway function in PPF populations. This study aims to evaluate the performance of several ML algorithms in diagnosing SAD in PPF population, and to find the best configuration.
Study population.
This was a single-centered, observational study in the Pulmonary Function Laboratory of West China Hospital, Sichuan University. Subjects were recruited and tested from May 1st to September 1st, 2020.
Included were adult patients undergoing pulmonary function tests as a result of persistent respiratory complaints. In addition, participants must meet the PPF requirements (FEV1/FVC ≥ 0.70) [ 4 ]. The following conditions had to be met in order to be excluded: restrictive pulmonary diseases (FVC < 80% predicted), asthma, interstitial lung diseases, lung cancer, respiratory infection within two weeks, myocardial ischemia, history of pulmonary surgery, and incomplete IOS due to tongue position errors, vocal cord closures, or swallowing. As healthy controls, we also enrolled never-smokers (those with ≤ 1 pack-year of tobacco smoking history) with a normal chest radiograph, no active pulmonary conditions, and no unstable cardiovascular disorders. Basic demographic data was gathered, such as height, weight, age, sex, and body mass index (BMI). Subjects received IOS, spirometry, and completed a questionnaire covering qualitative and quantitative evaluation of symptoms. Also, bronchial provocation tests or bronchodilator tests were performed to exclude asthma. The study was approved by the ethics committee of West China Hospital, Sichuan University, and all participants signed an informed consent before the procedure.
In accordance with ERS guidelines, the respiratory resistance and reactance were measured using IOS equipment (MS-IOS Jaeger) [ 9 ]. Because forced expiration may alter airway tone, IOS was performed prior to spirometry [ 13 ]. Pressure oscillations generated by a loudspeaker were superimposed onto normal tidal breathing through a mouthpiece for 30 to 45 s, which ranged from 5 to 35 Hz in frequency. Sitting upright, subjects were asked to wear a nasal clip and exert manual compression on their faces to minimize the influence of cheek vibration and air leak.
The IOS parameters selected in this paper and their clinical significance are as follows:
(1) Respiratory resistance at 5 Hz (R5): reflects the total viscous resistance of the respiratory system, because it is mainly airway resistance, also known as total airway resistance.
(2) Respiratory resistance at 20 Hz (R20): reflects central airway resistance.
(3) The difference between R5 and R20 (R5–R20): reflects the frequency dependence of resistance, that is, peripheral airway resistance. That is, the change of respiratory system resistance when the oscillation frequency is gradually increased.
(4) (R5-R20)/R5(%): the ratio of peripheral airway resistance to total airway resistance.
(5) Reactance at 5 Hz (X5): reflects the total elastic resistance of the respiratory system. Because the elastic resistance of the lung and thorax is the main one, it is often called peripheral elastic resistance, and also includes gas compression in the airway and alveoli. X5 is generally negative, with higher negative values indicating greater elastic resistance.
(6) Reactance area (AX): The area enclosed by the Xrs f frequency curve between 5 Hz and Fres and the horizontal 0 axis. AX is the integration of the low frequency reactance.
(7) Resonant frequency (Fres): The inertial resistance and elastic resistance are in opposite directions. When the two are equal and cancel each other, the reactance of the respiratory system is zero.
Spirometry was performed by a full MasterScreen PFT System (Jaeger Corp. Germany) according to the American Thoracic Society (ATS)/European Respiratory Society (ERS) guidelines [ 14 ]. FEV1, FVC, FEV1/ FVC, FEF25–75%, FEF50% and FEF75% were recorded as percentages of predicted values. The prediction equations are based on a large study of normal spirometry values in Chinese aged 4–80 years, which is recommended in the spirometry guideline in China [ 15 ].
The data collection used for the experiments included measurements from 280 participant groups. The data set contained information from the volunteers’ IOS test and lung function in addition to biological data like age, sex, height, and weight. The PPF patients without SAD (PPFN group) contributed 158 sets, the PPF patients with SAD (PPFA group) contributed 44 sets, and the healthy control group (CG group) contributed 78 sets. Using random sampling, the data set is split into training and test sets in a 7:3 ratio. All of the given results were from test sets. The adjustment of the hyperparameters was obtained by manual tuning, taking the hyperparameter with the best average result.
The discrete data measured by IOS can be thoroughly analyzed by ML algorithms to identify potential relationships. These ML algorithms were assessed in this study based on the findings of earlier research and pre-experiments:
(1) Random forests: A method of decision tree analysis in which a supervised algorithm works through “bagging” approach to create multiple decision trees with a random subset of the data. These decision trees are then merged to get a more accurate and stable prediction [ 16 ].
(2) Support vector machine: A supervised ML algorithm that classifies data points by finding the optimal hyperplane that maximally separates different classes in a high-dimensional space [ 17 ].
(3) Naive Bayes: A probabilistic classifier based on Bayes’ theorem [ 18 ].
(4) Adaptive Boosting (ADABOOST): A statistical classification algorithm that is frequently used with other “weaker” ML algorithms (e.g., decision tree) to improve their performance. [ 19 ]
(5) K-Nearest Neighbor (KNN): A common unsupervised ML method, in which unsupervised algorithms aim to group input vectors into k clusters based on k averages of points (i.e., centroids) without referring to known, or labeled outcomes [ 20 ].
In addition, this study conducted feature selection and investigated the use of SelectKBest, RFECV, and SelectFromModel algorithms in this experiment in order to find IOS parameters with a better correlation with the experimental results and minimize the complexity of the experimental data set.
(6) SelectKBest : A feature selection method based on statistical tests, which selects K features that are most relevant to the target variable according to some evaluation index. [ 21 ]
(7) RFECV: A Feature selection method in scikit-learn that combines Recursive Feature Elimination (RFE) and Cross-Validation (CV) to select the best feature subset [ 22 ].
(8) SelectFromModel: A feature selection method in scikit-learn, which selects the most relevant features based on the feature importance of the supervised learning model. [ 23 ]
This study involved the conduct of five experiments.
The first experiment’s goal was to assess each IOS parameter’s capacity to identify SAD in patients with PPF. The study’s criteria for diagnosing SAD were two out of the three small airway measurements (FEF25-75%, FEF50%, and FEF75%) having a predictive value of less than 65% according to spirometry. We examined two distinct scenarios: control versus PPF patients without SAD (CGvsPPFN) and control versus PPF patients with SAD (CGvsPPFA) in order to accurately assess the degree of airway blockage in patients with PPF. The two situations described were likewise assessed in the remaining studies.
The second experiment employed the ML algorithm and compared it to the results obtained using a single IOS parameter to ascertain whether the ML algorithm could achieve superior performance. The area under the ROC curve (AUC) was then selected as the performance evaluation metric. All IOS parameter characteristics for this experiment were included in the selection process.
In the third experiment, the effectiveness of SelectKBest as a feature selector for lowering complexity and determining the significance of various IOS parameters was evaluated. Five classifiers were used for training once SelectKBest had chosen the IOS parameters.
In the fourth and fifth experiments, two model-dependent feature selection algorithms were employed to investigate the significance of the 7 IOS feature parameters in this study.Recursive Feature Elimination with Cross-Validation, or RFECV, was used in Experiment 4. RFECV fits a machine learning model to data, ranks features according to their weights or importance, recursively removes the least important features, and uses cross-validation to assess model performance in each iteration. RFECV creates a performance curve by recording the results of varying numbers of features removed in each round. Using SelectFromModel, the most pertinent characteristics were chosen in Experiment 5 based on the significance of the features in a supervised learning model. To increase model efficiency and generalization while preserving important information, the technique selects features over a threshold, computes feature importance scores, trains a supervised learning model, and then generates a new feature set.
Hypothesis testing is necessary to contrast ML algorithms. A wide variety of parametric tests are available, often based on t-tests. The Wilcoxon Rank-Sum Test, the Kruskal-Wallis Test, and the Mann-Whitney U Test are a few of the most often used nonparametric tests [ 24 , 25 , 26 ]. We used the permutation test to do hypothesis testing of AUCs in this work. [ 27 , 28 ].
Table 1 displays the individuals’ biological parameters, spirometry results, chronic respiratory complaints, and IOS data. There was no discernible difference between any of the three research groups’ biological characteristics. There was no discernible difference in symptoms between the groups with and without spirometer-defined SAD for individuals with persistent respiratory symptoms. PPFA patients exhibited considerably lower spirometry parameters ( p < 0.05), as Table 1 illustrates.
(The last column describes the comparisons between groups, in which the dot means non-significant change, while the dash means significant change.)
Figure 1 ’s bar graphs display the distinct features of the IOS parameters for the CG, PPFN, and PPFA groups. The majority of IOS parameters were substantially different ( p < 0.05) across the three groups, according to the analysis of variance (ANOVA). PPF patients showed higher R5 and R20 when compared to healthy people. PPF patients consequently had greater airway resistance. In the meantime, patients with SAD in the PPF group showed greater values of R5, R5-R20, AXV, and Fres. The three groups’ R5-R20/R5 and X5 levels were comparable.
Comparison of IOS parameters among the three groups. Bar charts represented Mean + SD (M + SD). * indicates that there is a statistically significant difference comparing to each IOS parameter for each group. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001
The first experiment: diagnostic accuracy of IOS parameters.
Figure 2 presents the findings from Experiment 1. As can be observed, R5 was the best IOS parameter (BOP) for PPF patient diagnosis, with moderate diagnostic accuracy (AUC = 0.642, AUC = 0.769) for CG vs. PPFN and CG vs. PPFA scenarios.
Results of experiment 1, describing the diagnostic accuracy of Impulse oscillometry in subjects with chronic respiratory symptoms and preserved pulmonary function. More detailed tables and graphs regarding these results are available in the Additional file.(Additional file Figure S1 )
The second experiment of the study: diagnostic accuracy of the original IOS parameters associated with ML techniques.
Figure 3 presents the AUCs of the BOP, ML algorithm, and MIL classifier obtained in Experiment 2. It can be seen that the ML algorithm improves the AUC with high diagnostic accuracy in both cases, CGvsPPFN and CGvsPPFA. In the CGvsPPFN scenario, ADABOOST (AUC = 0.915) had the best performance, followed by RF (AUC = 0.914). Compared with BOP, RF, SVM, ADABOOST and KNN showed statistical differences. In the CGvsPPFA scenario, ADABOOST (AUC = 0.971) had the best performance, followed by RF (AUC = 0.951). Compared with BOP, RF, SVM, ADABOOST and KNN showed statistical differences.
Results of experiment 2, describing the diagnostic accuracy of Impulse oscillometry with ML algorithms in subjects with chronic respiratory symptoms and preserved pulmonary function. Also, * indicates that there a statistically significant difference comparing to BOP ( p < 0.05). * P < 0.05, ** P < 0.01. More detailed tables and graphs regarding these results are available in the Additional file.(Additional file Figure S2 - S3 )
The third experiment: diagnostic accuracy of the best original IOS parameters associated with ML techniques.
The IOS parameters used for the two cases, CGvsPPFN and CGvsPPFA, respectively, utilizing SelectKBest as the feature selector, are shown in Table 2 .
Experiments 2 and 3 had superior AUC outcomes, as shown by the data in Fig. 4 . A similar pattern was seen in both cases when SelectKBest was used as the feature selector: as the number of features increased, the ML algorithm’s performance improved over time. When choosing 3/5 IOS feature parameters, the AUC value decreased slightly, but overall, the diagnostic performance was still better than BOP.
Summary of Experiment 2 and Experiment 3 (SelectKBest as a feature selector)—AUCs for the best oscillometric parameter (BOP), for the best ML algorithms in experiments 3, and the best ML algorithm with oscillometric parameters (ADABOOST). The figure indicates the best ML algorithm in each case. Also, * indicates that there a statistically significant difference comparing to BOP ( p < 0.05). * P < 0.05, ** P < 0.01. More detailed tables and graphs regarding these results are available in the Additional file.(Additional file Figure S4 - S7 )
The fourth and fifth experiment: diagnostic accuracy of the IOS parameters associated with ML techniques.
The best AUC findings for Experiments 4 and 5 are shown in Fig. 5 . When compared to the full parameter, the IOS feature parameter’s diagnostic performance tends to be similar in both situations and to hold onto a high diagnostic value following feature selection.
The task configurations for each ML method classifier with the best performance across all experiments were summarized in Tables 3 and 4 . In the two scenarios of CGvsPPFN and CGvsPPFA, among them, RF, SVM, ADABOOST, and KNN may increase the AUC, and the difference was statistically significant. Furthermore, The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of various individual ML classifiers are also reported.
Summary of Experiment 4 and Experiment 5—AUCs for the best oscillometric parameter (BOP), for the best ML algorithms in experiments 4 and 5, and the best ML algorithm with oscillometric parameters. The figure indicates the best ML algorithm in each case. Also, * indicates that there a statistically significant difference comparing to BOP ( p < 0.05). * P < 0.05, ** P < 0.01.More detailed tables and graphs regarding these results are available in the Additional file.(Additional file Figure S8 - S19 )
For the purpose of early screening and treatment of respiratory disorders, a number of chronic respiratory disease guidelines, including GINA 2023 and GOLD 2024, advise early monitoring of changes in small airway function. In our previous study, we found that IOS is more sensitive to detect SAD than spirometry in subjects with chronic respiratory symptoms and PPF, and it correlates better with symptoms. IOS could be an additional method for SAD detection in the early stage of diseases [ 29 ]. Other similar research has demonstrated the usefulness of small airway function monitoring with IOS for clinical diagnosis [ 30 , 31 , 32 ]. We found only four correlated IOS parameters, including R5, R5-R20, AX, and Fres, which had low diagnostic efficacy, with none of the AUC values exceeding 0.7.
In order to facilitate the diagnosis of respiratory disorders, this study describes the design of a classifier for SAD diseases in the PPF population.By using machine learning approaches, this work aims to improve the diagnostic value of IOS for small airway dysfunction. Additionally, the best set of parameters and algorithms for this task was determined. Compared to a single IOS measure, the results show that this approach increases diagnostic accuracy and streamlines the clinical assessment of IOS.
Similar to our previous study, we found that R5 had the best AUC value, better sensitivity and slightly lower specificity among all parameters. After the introduction of the machine learning algorithm, the AUC, sensitivity, and specificity of the prediction model were very significantly improved.The best performance in both CGvsPPFN and CGvsPPFA scenarios was achieved by R5, which was the single IOS parameter used in the first experiment. The finding supports the presence of elevated airway resistance in patients with SAD, as measured by various methods including CT scans and bronchoscopy. It is important to note that these results are based on objective measurements rather than subjective evaluations [ 33 , 34 ].
In the first case, it was more challenging to differentiate the control group from the patients with PPF who had preserved lung function. This was due to the small differences in IOS parameters. The AUC value was 0.642, indicating low diagnostic accuracy. In the second case, the increase in physiological abnormalities resulted in a greater difference in measured parameters, enabling R5 to easily distinguish between the two groups with an AUC of 0.769. These findings suggest that a single IOS parameter may not be sufficient to accurately identify the SAD situation in the PPF population.
The diagnostic accuracy was significantly enhanced through the utilization of RF, SVM, BAYES, ADABOOST, and KNN algorithms. It is clear that ADABOOST and RF produced the most favorable results followed by KNN, SVM and BYS.This breakthrough is mainly due to the use of ML algorithms.Similar to earlier research [ 35 , 36 , 37 , 38 ], feature selection permits the use of fewer characteristics without appreciably lowering performance. When SelectKBest was employed as a feature selector, the 3/5 relevant features were selected, respectively. Despite the final trend indicating that the results are superior when more parameters are used, the difference between using the least and most parameters is relatively minor. Furthermore, the results are superior when using the least parameters than when using BOP alone. This implies that feature selection can in fact result in good diagnostic value (AUC 0.948 and 0.967, respectively) with fewer IOS parameters. The most pertinent features are found through feature selection in both the CGvsPPFN and CGvsPPFA scenarios. Despite the fact that the approach only chose two sets of features, R20 and Fres had a significant intersection. This intersection is slightly different from the results of the ability of each single IOS parameter to diagnose SAD in patients with PPF, showing better diagnostic ability for R5 when using a single parameter. This suggests resonant frequency and central airway resistance, in addition to total airway resistance, have a significant role in the increased airway blockage observed in the PPF population.
Compared to the conventional classifier SelectFromModel, the RFECV method may produce superior results and has an efficient selection capability. While it does not increase the accuracy of diagnosis, it does display significant traits like R5, (R5-R20)/R5, and Fres. Feature selection was done to make the analysis easier to understand. We were able to discriminate between groups with clarity by using these three essential criteria. These results support the idea of a simple diagnostic model that can help explain the suggested medical decision support system’s findings and make it easier to apply in clinical settings.
Recent studies have shown that IOS is considered the most advanced technique for lung function analysis and is one of the most promising emerging techniques in the field [ 29 , 39 , 40 , 41 ]. Despite its advantages in providing detailed and direct examination, IOS has not yet been widely used. However, because interpreting the metrics—which are based on electrical modeling—requires knowledge and experience, their application is restricted. This study shows how ML algorithms can improve the diagnosis of associated diseases and simplify the use of IOS, therefore improving healthcare for patients with SAD.
Early detection of abnormal respiratory changes in SAD can facilitate timely interventions that may limit disease progression, alleviate adverse symptoms, improve overall health, prevent complications and comorbidities, and reduce premature mortality [ 5 , 42 ]. Since the 1980s, lung function analysis has been improved by artificial intelligence and machine learning techniques [ 43 , 44 , 45 , 46 , 47 , 48 ]. The present work expands on previous results by demonstrating that early aberrant respiratory alterations in SAD may be suggested by a combination of IOS measures and a clinical decision support system based on ML technology.
The algorithm presented in this work can be applied not just to SAD but to a variety of other conditions, including asthma, COPD, interstitial lung disease, and others. By establishing appropriate models and finding the best parameters, the relationship between physiological parameters and the development of the disease can be explored. This benefits the early screening of other respiratory diseases and the reduction of the disease burden on patients.
Clinical technology-wise, more thorough information can be obtained by combining IOS with other imaging modalities (such as MRI, CT, PET, etc.) and by developing real-time imaging technology and dynamic observation techniques. More information for clinical diagnosis and scientific study will be available with the improvement of image contrast and anatomical detail. [ 49 ] Concurrently, artificial intelligence and machine learning are integrated to analyse and interpret multiple data types, enhance the accuracy and credibility of clinical examination results, and develop automated and intelligent analysis tools. Encouraging data sharing and IOS standardization, creating a platform for data sharing and standardizing data formats, facilitating multi-center data comparison and analysis, and promoting the field’s progress are all crucial in the context of big data [ 50 ].
Finally, it is important to consider and clarify some significant limitations. Firstly, this study is limited to the Chinese population in a specific location. Therefore, it is not possible to ensure its generalisability to different populations. It is recommended that future studies investigate multi-centre data to expand the generalisability of the findings. The experimental design of this work followed globally recognised inclusion and exclusion criteria and was conducted in a typical clinical setting.
Additionally, it is important to note that the PPF population in China is relatively small due to low public health awareness. Many individuals do not seek medical attention promptly when experiencing clinical symptoms such as cough and chest tightness. Therefore, due to the relatively small size of the available dataset, it is necessary to carefully control the complexity of the ML model. In addition to the measures taken in this study to avoid overfitting, such as controlling hyperparameters, feature selection can also aid in controlling overfitting by reducing inputs. Another reason for using feature selection is that a smaller number of features can help simplify the analysis. Furthermore, utilising only three features enables the visualisation of group separation, aiding diagnostic interpretation.
In this work, a variety of machine learning algorithms were utilized to create a clinical auxiliary diagnosis system that can identify respiratory anomalies in patients with PPF. In the initial disease stage (CGvsPPFN), respiratory oscillation parameters achieved low diagnostic accuracy (AUC = 0.642), but ML classifiers significantly improved accuracy (AUC ≥ 0.9). In the progressive disease stage (CGvsPPFA), using oscillation parameters alone yielded moderate accuracy (AUC = 0.769), while ML algorithms greatly enhanced accuracy (AUC ≥ 0.9). The developed diagnostic system simplifies IOS application in PPF patients, utilizing key IOS parameters identified through feature selection. All things considered, combining ML algorithms with IOS examination improves pulmonary function assessment in PPF patients, indicating future improvements in patient care.
No datasets were generated or analysed during the current study.
Chronic obstructive pulmonary disease
Random Forests
Support Vector Machines
Navie Bayesian
Adaptive Boosting
K-Nearest Neighbors
The forced expiratory volume in 1st s
Forced vital capacity
Airway hyper-responsiveness
Bronchial reversibility
The forced expiratory flow between 25 and 75% of FVC; FEF50%:The forced expiratory flow when 50% of FVC has been exhaled
The forced expiratory flow when 75% of FVC has been exhaled
Resistance of the respiratory system
Reactance of the respiratory system
Respiratory resistance at 5 Hz
Respiratory resistance at 20 Hz
The difference between R5 and R20
Reactance at 5 Hz
Resonant Frequency
Area under reactance curve between Fres and5 Hz
Best Oscillometric Parameter
American Thoracic Society
Europe Respiratory Society
Receiver Operator Characteristic
Area Under the Curve
Positive Predictive Value
Negative Predictive Value
Body Mass Index
Stockley JA, Ismail AM, Hughes SM, Edgar R, Stockley RA, Sapey E. Maxi–Mal mid-expiratory flow detects early lung disease in α(1)-antitrypsin deficiency. Eur Respir J. 2017;49:1602055.
Article PubMed Google Scholar
Schroeder JD, McKenzie AS, Zach JA, Wilson CG, Curran-Everett D, Stinson DS, Newell JD Jr, Lynch DA. Relationships between airflow obstruction and quantitative CT measurements of emphysema, air trapping, and airways in subjects with and without chronic obstructive pulmonary disease. AJR Am J Roentgenol. 2013;201:W460–470.
Article PubMed PubMed Central Google Scholar
Skylogianni E, Triga M, Douros K, Bolis K, Priftis KN, Fouzas S. Anthraco–Poulos MB. Small-airway dysfunction precedes the development of asthma in children with allergic rhinitis. Allergol Immunopathol (Madr). 2018;46:313–21.
Article CAS PubMed Google Scholar
Woodruff PG, Barr RG, Bleecker E, Christenson SA, Couper D, Curtis JL, Gouskova NA, Hansel NN, Hoffman EA, Kanner RE, et al. Clinical significance of symptoms in smokers with preserved pulmonary function. NEngl J Med. 2016;374:1811–21.
Article CAS Google Scholar
Xiao D, Chen Z, Wu S, Huang K, Xu J, Yang L, Xu Y, Zhang X, Bai C, Kang J, et al. Prevalence and risk factors of small airway dysfunction, and association with smoking, in China: findings from a national cross-sectional study. Lancet Respir Med. 2020;8:1081–93.
Burgel PR, Bergeron A, de Blic J, et al. Small airways diseases, excluding asthma and COPD: an overview. Eur Respir Rev. 2013;22(128):131–47. https://doi.org/10.1183/09059180.00001313 .
Konstantinos Katsoulis K, Kostikas K, Kontakiotis T. Techniques for assessing small airways function: possible applications in asthma and COPD. Respir Med. 2016;119:e2–9.
Contoli M, Bousquet J, Fabbri LM, Magnussen H, Rabe KF, Siafakas NM, Hamid Q, Kraft M. The small airways and distal lung compartment in asthma and COPD: a time for reappraisal. Allergy. 2010;65:141–51.
King GG, Bates J, Berger KI, Calverley P, de Melo PL, Dellaca RL, Farre R, Hall GL, Ioan I, Irvin CG, et al. Technical standards for respiratory oscillometry. Eur Respir J. 2020;55:1900753.
Skloot G, Goldman M, Fischler D, Goldman C, Schechter C, Levin S, Teirstein A. Respiratory symptoms and physiologic assessment of ironworkers at the World Trade Center disaster site. Chest. 2004;125:1248–55.
Oppenheimer BW, Goldring RM, Herberg ME, Hofer IS, Reyfman PA, Liautaud S, Rom WN, Reibman J, Berger KI. Distal airway function in symptomatic subjects with normal spirometry following World Trade Center dust exposure. Chest. 2007;132:1275–82.
Su ZQ, Guan WJ, Li SY, Ding M, Chen Y, Jiang M, Chen XB, Zhong CH, Tang CL, Zhong NS. Significances of spirometry and impulse oscillometry for detecting small airway disorders assessed with endobronchial optical coherence tomography in COPD. Int J Chron Obstruct Pulmon Dis. 2018;13:3031–44.
Williamson PA, Clearie K, Menzies D, Vaidyanathan S, Lipworth BJ. Assessment of small-airways disease using alveolar nitric oxide and impulse oscillometry in asthma and COPD. Lung. 2011;189:121–9.
Graham BL, Steenbruggen I, Miller MR, et al. Standardization of spirometry 2019 update. An official American thoracic society and European respiratory society technical statement[J]. Am J Respir Crit Care Med. 2019;200(8):e70–88.
Jian W, Gao Y, Hao C, et al. Reference values for spirometry in Chinese aged 4–80 years[J]. J Thorac Disease. 2017;9(11):4538.
Article Google Scholar
BREIMAN L. Random forests. ML, 2001, 45: 5–32.
CORTES C. Vladimir. Support-vector networks. ML. 1995;20:273–97.
Google Scholar
MANNING, RAGHAVAN CD. Prabhakar; SCHÜTZE, Hinriche. Xml retrieval. Introduction to Information Retrieval; 2008.
SCHAPIRE FREUNDY, Robert E. A decision-theoretic generalization of on-line learning and an application to boosting. J Comput Syst Sci. 1997;55(1):119–39.
COVER T, HART P. Nearest neighbor pattern classification. IEEE Trans Inf Theory. 1967;13(1):21–7.
PEDREGOSA F, et al. Scikit-learn: ML in Python. J ML Res. 2011;12:2825–30.
GUYON, Isabelle, et al. Gene selection for cancer classification using support vector machines. ML. 2002;46:389–422.
GUYON I, ELISSEEFF André. An introduction to variable and feature selection. J ML Res. 2003;3:1157–82.
WILCOXON F. Individual comparisons by ranking methods. Breakthroughs in statistics: methodology and distribution. New York, NY: Springer New York; 1992. pp. 196–202.
Chapter Google Scholar
WALLIS KRUSKALWH, Allen W. Use of ranks in one-criterion variance analysis. J Am Stat Assoc, 1952, 583–621.
MANN, WHITNEY HB, Donald R. On a test of whether one of two random variables is stochastically larger than the other. Ann Math Stat, 1947, 50–60.
Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143(1):29–36.
DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach[J]. Biometrics, 1988: 837–45.
Li LY, Yan TS, Yang J, et al. Impulse oscillometry for detection of small airway dysfunction in subjects with chronic respiratory symptoms and preserved pulmonary function[J]. Respir Res. 2021;22:1–10.
Chiu HY, Hsiao YH, Su KC, Lee YC, Ko HK, Perng DW. Small Airway Dysfunction by Impulse Oscillometry in symptomatic patients with preserved pulmonary function. J Allergy Clin Immunol Pract. 2020;8(1):229–e2353. https://doi.org/10.1016/j.jaip.2019.06.035 .
Crisafulli E, Pisi R, Aiello M, et al. Prevalence of small-airway dysfunction among COPD patients with different GOLD stages and its role in the impact of Disease. Respiration. 2017;93(1):32–41. https://doi.org/10.1159/000452479 .
Anderson WJ, Zajda E, Lipworth BJ. Are we overlooking persistent small airways dysfunction in community-managed asthma? Ann Allergy Asthma Immunol. 2012;109(3):185–e1892. https://doi.org/10.1016/j.anai.2012.06.022 .
Postma DS, Brightling C, Baldi S, et al. Exploring the relevance and extent of small airways dysfunction in asthma (ATLANTIS): baseline data from a prospective cohort study[J]. Lancet Respiratory Med. 2019;7(5):402–16.
McNulty W, Usmani OS. Techniques of assessing small airways dysfunction[J]. Eur Clin Respiratory J. 2014;1(1):25898.
Amaral JLM, Sancho AG, Faria ACD, Lopes AJ, Melo PL. Differential diagnosis of asthma and restrictive respiratory diseases by combining forced oscillation measurements, machine learning and neuro-fuzzy classifiers. Med Biol Eng Comput. 2020;58(10):2455–73.
Amaral JLM, Lopes AJ, Jansen JM, Faria ACD, Melo PL. Machine learning algorithms and forced oscillation measurements applied to the automatic identification of chronic obstructive pulmonary disease. Comput Methods Programs Biomed. 2012;105(3):183–93.
Amaral JLM, Lopes AJ, Veiga J, et al. High-accuracy detection of airway obstruction in asthma using machine learning algorithms and forced oscillation measurements[J]. Comput Methods Programs Biomed. 2017;144:113–25.
Andrade DSM, Ribeiro LM, Lopes AJ, et al. Machine learning associated with respiratory oscillometry: a computer-aided diagnosis system for the detection of respiratory abnormalities in systemic sclerosis[J]. Biomed Eng Online. 2021;20(1):1–18.
Bednarek M, Grabicki M, Piorunek T, et al. Current place of impulse oscillometry in the assessment of pulmonary diseases[J]. Respir Med. 2020;170:105952.
Sarkar S, Jadhav U, Ghewade B et al. Oscillometry in lung function Assessment: a Comprehensive Review of Current insights and Challenges[J]. Cureus, 2023, 15(10).
Avila N, Nazeran H, Gordillo N, et al. Computer-aided classification of small airways dysfunction using impulse oscillometric features: a children-focused review[J]. Biomedical Engineering/Biomedizinische Technik. 2020;65(2):121–31.
Cottini M, Lombardi C, Berti A, et al. Small-airway dysfunction in paediatric asthma[J]. Curr Opin Allergy Clin Immunol. 2021;21(2):128–34.
Topalovic M, Das N, Burgel PR et al. Artificial intelligence outperforms pulmonologists in the interpretation of pulmonary function tests[J]. Eur Respir J, 2019, 53(4).
Das N, Happaerts S, Gyselinck I et al. Collaboration between explainable artificial intelligence and pulmonologists improves the accuracy of pulmonary function test interpretation[J]. Eur Respir J, 2023, 61(5).
Giri PC, Chowdhury AM, Bedoya A, et al. Application of machine learning in pulmonary function assessment where are we now and where are we going?[J]. Front Physiol. 2021;12:678540.
Wang Y, Li Y, Chen W, et al. Deep learning for spirometry quality assurance with spirometric indices and curves[J]. Respir Res. 2022;23(1):1–9.
Das N, Verstraete K, Stanojevic S et al. Deep-learning algorithm helps to standardise ATS/ERS spirometric acceptability and usability criteria[J]. Eur Respir J, 2020, 56(6).
Park H, Yun J, Lee SM, et al. Deep learning–based Approach to predict pulmonary function at chest CT[J]. Radiology. 2023;307(2):e221488.
Wichum F, Wiede C, Seidl K. Depth-based measurement of respiratory volumes: a Review[J]. Sensors. 2022;22(24):9680.
Article CAS PubMed PubMed Central Google Scholar
Wang Y, Li Q, Chen W, et al. Deep learning-based analytic models based on flow-volume curves for identifying ventilatory patterns[J]. Front Physiol. 2022;13:824000.
Download references
Not applicable.
This study was partly supported by the National Nature Science Foundation of China Grant (NSFC No.81800016), Sichuan Science and Technology Agency Grant (2019YFS0033). The funders had no roles in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Authors and affiliations.
Department of Respiratory and Critical Care Medicine, West China School of Medicine and West China Hospital, Sichuan University, No. 37 Guoxue Alley, Chengdu, 610041, China
Wen-Jing Xu, Xin-Yue Song, Liang-Yuan Li & Bin-Miao Liang
Machine Intelligence Laboratory, College of Computer Science, Sichuan University, Chengdu, 610065, China
Wen-Yi Shang
West China School of Medicine, West China Hospital, Sichuan University, Chengdu, Sichuan, P. R. China
Jia-Ming Feng
College of Electrical Engineering and Automation, Sichuan University, Chengdu, 610065, China
Xin-Peng Xie
Institute of Traditional Chinese Medicine of Sichuan Academy of Chinese Medicine Sciences(Sichuan Second Hospital of T.C.M), Chengdu, 610000, China
Yan-Mei Wang
You can also search for this author in PubMed Google Scholar
XWJ contributed to study design, manuscript writing and data analysis. SWY contributed to data acquisition and analysis. SXY and LLY contributed to study design and data interpretation. XXP and WYM contributed to data acquisition and interpretation. LBM contributed to study design and manuscript revision. All authors Read and approved the final manuscript. FJM contributed to the linguistic embellishment of the article as well as proofreading of the manuscript.
Correspondence to Bin-Miao Liang .
Ethical approval and consent to participate.
This study conformed to the Declaration of Helsinki and was approved by the Ethics Committee of West China Hospital, Sichuan University, China. All participants signed an informed consent before the procedure.
Competing interests.
The authors declare no competing interests.
Publisher’s note.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Below is the link to the electronic supplementary material.
Rights and permissions.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Reprints and permissions
Cite this article.
Xu, WJ., Shang, WY., Feng, JM. et al. Machine learning for accurate detection of small airway dysfunction-related respiratory changes: an observational study. Respir Res 25 , 286 (2024). https://doi.org/10.1186/s12931-024-02911-1
Download citation
Received : 06 March 2024
Accepted : 08 July 2024
Published : 24 July 2024
DOI : https://doi.org/10.1186/s12931-024-02911-1
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
ISSN: 1465-993X
You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.
All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .
Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.
Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.
Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.
Original Submission Date Received: .
Find support for a specific problem in the support section of our website.
Please let us know what you think of our products and services.
Visit our dedicated information section to learn more about MDPI.
A method to obtain frequency response functions of operating mechanical systems based on experimental modal analysis and operational modal analysis.
2. problem description and the main assumption, 3. the proposed method, 3.1. the main idea and procedures, 3.2. the principal square root method for the third step, 3.3. the stability of condensed mass and countermeasures, 3.4. practical implementation.
4.1. simulation example, 4.2. experimental example, 5. conclusions, supplementary materials, author contributions, data availability statement, acknowledgments, conflicts of interest.
Click here to enlarge figure
Node | Coordinate | Node | Coordinate |
---|---|---|---|
1001 | (1000.0, 0.0, 0.0) | 1011 | (0.0, 480.0, 0.0) |
1002 | (1000.0, 210.0, 0.0) | 1012 | (0.0, 180.0, 0.0) |
1003 | (1000.0, 490.0, 0.0) | 1013 | (0.0, 0.0, 0.0) |
1004 | (1000.0, 780.0, 0.0) | 1014 | (240.0, 0.0, 0.0) |
1005 | (1000.0, 1000.0, 0.0) | 1015 | (510.0, 0.0, 0.0) |
1006 | (760.0, 1000.0, 0.0) | 1016 | (790.0, 0.0, 0.0) |
1007 | (480.0, 1000.0, 0.0) | 1017 | (1000.0, 520.0, 0.0) |
1008 | (180.0, 1000.0, 0.0) | 1018 | (520.0, 1000.0, 0.0) |
1009 | (0.0, 1000.0, 0.0) | 1019 | (500.0, 1000.0, 0.0) |
1010 | (0.0, 790.0, 0.0) | 1020 | (0.0, 500.0, 0.0) |
The Original Structure | The Targeted Structure |
---|---|
11.3829 | 10.8636 |
19.7494 | 18.7843 |
25.0262 | 19.6450 |
36.8577 | 23.6460 |
43.3347 | 35.4094 |
52.3165 | 42.5321 |
53.6780 | 50.3279 |
57.9709 | 52.1611 |
71.4442 | 55.4953 |
83.3543 | 68.4422 |
89.5122 | 79.8620 |
99.1503 | 85.4602 |
102.9488 | 93.4075 |
125.8974 | 98.0684 |
132.2300 | 119.3317 |
138.6529 | 125.5628 |
152.4326 | 132.6460 |
163.3064 | 146.4351 |
168.6131 | 154.5092 |
188.2306 | 161.2042 |
218.2984 | 178.4255 |
223.2421 | 209.2524 |
229.6832 | 213.1208 |
The Original Structure | The Targeted Structure |
---|---|
11.3825 | 12.5928 |
19.7554 | 21.4307 |
25.0330 | 27.2876 |
36.8915 | 39.8742 |
43.6217 | 57.7508 |
52.6610 | 62.9120 |
53.7115 | 77.2842 |
57.9793 | 82.8056 |
71.4567 | 96.0678 |
83.3827 | 97.8849 |
89.4947 | 110.8344 |
99.1247 | 112.9849 |
102.9327 | 137.5893 |
125.8468 | 144.9645 |
132.2340 | 151.1522 |
138.5813 | 162.2044 |
152.3783 | 179.5252 |
163.3391 | 184.6468 |
168.6119 | 207.0478 |
The Original Structure | The Targeted Structure |
---|---|
56.2318 | 57.9509 |
64.9327 | 64.2205 |
109.6792 | 115.5228 |
154.3924 | 164.5742 |
201.9135 | 223.7241 |
232.8787 | 237.3319 |
260.2597 | 263.0481 |
284.6582 | 310.9808 |
319.4749 | 329.2040 |
333.7813 | 350.3971 |
372.3621 | 361.5300 |
405.0700 | 388.7280 |
420.2550 | 412.9673 |
443.1399 | 436.9488 |
Items | Original System | Unit |
---|---|---|
Coordinate of P | (300, 500, 300) | mm |
Side length of plate | 1000 | mm |
Thickness of plate | 5 | mm |
Section diameter of beam | 6 | mm |
Material density | kg/mm | |
Elasticity modulus | 210 | GPa |
Poisson’s ratio | 0.3 | — |
Mass of mass point | 1 | kg |
Stiffness of spring | 0.5, 100, 100 | N/mm |
Damping of spring | 5, 100, 100 | N·s/mm |
Shell element size | mm | |
Number of elements per beam | 1 | — |
DOF | The OMA Result | The Exact Mode Shape |
---|---|---|
A | 0.0525 − 0.1887i | −0.0623 |
B | 0.1104 − 0.0493i | 0.0705 |
C | −0.0585 + 0.0010i | −0.0590 |
D | 1 | 1 |
O | 0.4604 + 0.0466i | 0.4771 |
The lumped mass | 0.1796 − 0.0151i | 0.1676 |
The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
Shen, C.; Lu, C. A Method to Obtain Frequency Response Functions of Operating Mechanical Systems Based on Experimental Modal Analysis and Operational Modal Analysis. Machines 2024 , 12 , 516. https://doi.org/10.3390/machines12080516
Shen C, Lu C. A Method to Obtain Frequency Response Functions of Operating Mechanical Systems Based on Experimental Modal Analysis and Operational Modal Analysis. Machines . 2024; 12(8):516. https://doi.org/10.3390/machines12080516
Shen, Cunrui, and Chihua Lu. 2024. "A Method to Obtain Frequency Response Functions of Operating Mechanical Systems Based on Experimental Modal Analysis and Operational Modal Analysis" Machines 12, no. 8: 516. https://doi.org/10.3390/machines12080516
Article access statistics, supplementary material.
ZIP-Document (ZIP, 105289 KiB)
Mdpi initiatives, follow mdpi.
Subscribe to receive issue release notifications and newsletters from MDPI journals
This work presents an experimental realization of a ring scheme of nonlinear threshold controlled unidirectionally coupled (N = 3) second-order autonomous type oscillating systems. The originality of this work lies in having the threshold controller as the nonlinear element of the dynamical system and as the coupling element to form a ring circuit using these systems. The advantage of this coupling is getting tuning of frequency (multi-frequency) of the ring from a few hertz to kilohertz along with the observation of a periodic rotating wave pattern by varying one of the parameter values of the system, in terms of either changing the resistor value (gain) in the coupling path or changing the threshold value of the threshold controller or both. The results explored through this experimental study are confirmed by numerically simulated results, obtained using MATLAB coding- simulink and MULTISIM software. The symmetrical and asymmetrical aspects of the flexible threshold coupling are also studied and the observed interesting experimental and numerical results are presented.
This is a preview of subscription content, log in via an institution to check access.
Subscribe and save.
Price includes VAT (Russian Federation)
Instant access to the full article PDF.
Rent this article via DeepDyve
Institutional subscriptions
The datasets generated during and analysed during the current study are not publicly available from the corresponding author on reasonable request.
D. Armbruster, P. Chossat, Remarks on multi-frequency oscillations in symmetrically coupled oscillators. Phys. Lett. A. 254 (5), 269–274 (1999)
Article Google Scholar
S. Balamurali, G. Mangraviti, C.-H. Tsai, P. Wambacq, J. Craninckx, Design and analysis of 55–63-GHz fundamental quad-core VCO with NMOS-only stacked oscillator in 28-nm CMOS. IEEE J. Solid-State Circuits 57 (7), 1997–2010 (2022)
A.R. Bulsara, V. In, A. Kho, P. Longhini, A. Palacios, W.-J. Rappel, J. Acebron, S. Baglio, B. Ando, Emergent oscillations in unidirectionally coupled overdamped bistable systems. Phys. Rev. E. 70 (3), 036103 (2004)
A. Chithra, I. Raja Mohamed, Synchronization and chaotic communication in nonlinear circuits with nonlinear coupling. J. Comput. Electron. 16 (3), 833–844 (2017)
H. Fujisaka, T. Yamada, Stability theory of synchronized motion in coupled-oscillator systems. Progr. Theor. Phys. 69 (1), 32–47 (1983)
Article MathSciNet Google Scholar
Y. Horikawa, H. Kitajima, Transient chaotic rotating waves in a ring of unidirectionally coupled symmetric Bonhoeffer-van der Pol oscillators near a codimension-two bifurcation point. Chaos Interdiscip. J. Nonlinear Sci. 22 (3), 32–47 (2012)
V. In, A. Kho, J.D. Neff, A. Palacios, P. Longhini, B.K. Meadows, Experimental observation of multifrequency patterns in arrays of coupled nonlinear oscillators. Phys. Rev. Lett. 91 (24), 244101 (2003)
V. In, P. Longhini, A. Kho, N. Liu, S. Naik, A. Palacios, J.D. Neff, Frequency down-conversion using cascading arrays of coupled nonlinear oscillators. Phys. D Nonlinear Phenom. 240 (8), 701–708 (2011)
V. In, P. Longhini, A. Kho, B.K. Meadows, A. Palacios, Nonlinear channelizer: a frequency agile and adaptive receiver for RF communication. IEICE Proc. Ser. 2 (106), 106–109 (2013)
Google Scholar
V. In, A. Palacios, A.R. Bulsara, P. Longhini, A. Kho, J.D. Neff, S. Baglio, B. Ando, Complex behavior in driven unidirectionally coupled overdamped duffing elements. Phys. Rev. E. 73 (6), 066121 (2006)
V. In, A. Palacios, Dynamics and bifurcations in networks designed for frequency conversion. Int. J. Bifurc. Chaos 31 (12), 2130037 (2021)
A. Kerber, T. Nigam, P. Paliwoda, F. Guarin, Reliability characterization of ring oscillator circuits for advanced CMOS technologies. IEEE Trans. Device Mater. Rel. 20 (2), 230–241 (2020)
P. Longhini, A. Palacios, V. In, J.D. Neff, A. Kho, A. Bulsara, Exploiting dynamical symmetry in coupled nonlinear elements for efficient frequency down-conversion. Phys. Rev. Lett. 76 (2), 026201 (2007)
J. Lü, K. Murali, S. Sinha, H. Leung, M. Aziz-Alaoui, Generating multi-scroll chaotic attractors by thresholding. Phys. Lett. A. 372 (18), 3234–3239 (2008)
M.A. Matías, V. Pérez-Muñuzuri, M.N. Lorenzo, I.P. Mariño, V. Pérez-Villar, Observation of a fast rotating wave in rings of coupled chaotic oscillators. Phys. Rev. Lett. 78 (2), 219–222 (1997)
K. Murali, M. Lakshmanan, Drive-response scenario of chaos synchronization in identical nonlinear systems. Phys. Rev. E. 49 (6), 4882 (1994)
T. Oguchi, T. Yamamoto, H. Nijmeijer, Synchronisation of nonlinear systems by bidirectional coupling with time-varying delay. IFAC Proc. Vol. 40 (23), 262–267 (2007)
A. Palacios, R. Carretero-González, P. Longhini, N. Renz, V. In, A. Kho, J.D. Neff, B.K. Meadows, A.R. Bulsara, Multifrequency synthesis using two coupled nonlinear oscillator arrays. Phys. Rev. E. 72 (2), 026211 (2005)
P. Perlikowski, S. Yanchuk, M. Wolfrum, A. Stefanski, P. Mosiolek, T. Kapitaniak, Routes to complex dynamics in a ring of unidirectionally coupled systems. Chaos Interdiscip. J. Nonlinear Sci. 20 (1), 013111 (2010)
I. Raja Mohamed, K. Murali, S. Sinha, E. Lindberg, Design of threshold controller based chaotic circuits. Int. J. Bifurc. Chaos 20 (7), 2185–2191 (2010)
E. Sánchez, M.A. Matías, Transition to chaotic rotating waves in arrays of coupled Lorenz oscillators. Int. J. Bifurc. Chaos 9 (12), 2335–2343 (1999)
E. Sánchez, M.A. Matías, Experimental observation of a periodic rotating wave in rings of unidirectionally coupled analog Lorenz oscillators. Phys. Rev. E. 57 (5), 6184–6186 (1998)
G.-Y. Su, S.-I. Liu, A 1.22 mW 2.4 GHz PLL using a single-ring-oscillator-based integrator with background frequency calibration. IEEE Trans. Circuits Syst. I: Regul. Pap. 67 (7), 2169–2179 (2020)
C.K. Volos, I.M. Kyprianidis, I.N. Stouboulos, Image encryption process based on chaotic synchronization phenomena. Signal Process. 93 (5), 1328–1340 (2013)
H. Zhang, S. Li, T. Iizuka, A single ring-oscillator-based test structure for timing characterization of dynamic circuit. IEEE Trans. Very Large Scale Integr. (VLSI) Syst. 32 (5), 938–951 (2024)
Download references
The authors would like to thank Dr. K. Murali, Department of Physics, Anna University, Chennai, for his valuable guidance and support and Dr. A. Abudhahir, Department of EIE, BSA Crescent Institute of Science and Technology, Chennai, for the fruitful discussion in completing this work. P. Yogamarish acknowledge B. S. Abdur Rahman Crescent Institute of Science and Technology, Chennai-600 048, for the award of BSA - Junior Research Fellowship (File No.: Lr. No.475 / Dean (R) / 2023).
The authors declare that they did not receive any financial assistance, grands, or support while preparing this manuscript.
Authors and affiliations.
Department of Physics, B. S. Abdur Rahman Crescent Institute of Science and Technology, Chennai, Tamil Nadu, 600048, India
P. Yogamarish & I. Raja Mohamed
You can also search for this author in PubMed Google Scholar
All authors contributed equally to the study conception, design, material preparation, data collection and analysis were performed by P. Yogamarish and I. Raja Mohamed. The first draft of the manuscript was prepared by P. Yogamarish under the supervision of I. Raja Mohamed and the final draft of the manuscript was approved by the corresponding author (I. Raja Mohamed).
Correspondence to I. Raja Mohamed .
Conflict of interest.
The authors have no relevant financial or non-financial interest to disclose.
Publisher's note.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
Reprints and permissions
Yogamarish, P., Mohamed, I.R. Multi-frequency Oscillations in the Nonlinear Threshold Controlled Unidirectionally Coupled Oscillators. Circuits Syst Signal Process (2024). https://doi.org/10.1007/s00034-024-02795-y
Download citation
Received : 14 March 2024
Revised : 11 July 2024
Accepted : 11 July 2024
Published : 29 July 2024
DOI : https://doi.org/10.1007/s00034-024-02795-y
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
COMMENTS
For example, studying the long-term effects of smoking requires an observational approach because we can't ethically assign people to smoke or abstain from smoking. Cost-Effective: Observational studies are generally less expensive and time-consuming than experiments. Longitudinal Research: They are well-suited for long-term studies or those ...
Observational study. Experiment. B. Experiment. Check. Another study took a group of adults and randomly divided them into two groups. One group was told to drink tea every night for a week, while the other group was told not to drink tea that week. Researchers then compared when each group fell asleep. question b.
Observational research studies involve the passive observation of subjects without any intervention or manipulation by researchers. These studies are designed to scrutinize the relationships between variables and test subjects, uncover patterns, and draw conclusions grounded in real-world data. Researchers refrain from interfering with the ...
There are seven types of observational studies. Researchers might choose to use one type of observational study or combine any of these multiple observational study approaches: 1. Cross-sectional studies. Cross-sectional studies happen when researchers observe their chosen subject at one particular point in time.
For example, if they want to study the effects of a disease, they can't give a population the disease to perform the experiment. In that case, it is better to conduct an observational study. In observational studies, researchers observe people or objects in their natural state. There is no interference, so this type of study is usually easier.
Observational Study Definition. In an observational study, the researchers only observe the subjects and do not interfere or try to influence the outcomes. In other words, the researchers do not control the treatments or assign subjects to experimental groups. Instead, they observe and measure variables of interest and look for relationships ...
An observational study is used to answer a research question based purely on what the researcher observes. There is no interference or manipulation of the research subjects, and no control and treatment groups. These studies are often qualitative in nature and can be used for both exploratory and explanatory research purposes.
From an epidemiological standpoint, there are two major types of clinical study designs, observational and experimental. 3 Observational studies are hypothesis‐generating studies, and they can be further divided into descriptive and analytic. Descriptive observational studies provide a description of the exposure and/or the outcome, and ...
Observational versus Experimental Studies# In most research questions or investigations, ... An example of a retrospective study would be using administrative data from a hospital to study incidence of disease. In contrast, a prospective study identifies subjects beforehand and collects data as events unfold. For example, one might use a ...
A study where a researcher records or observes the observations or measurements without manipulating any variables. These studies show that there may be a relationship but not necessarily a cause and effect relationship. Experimental. A study that involves some random assignment* of a treatment; researchers can draw cause and effect (or causal ...
In observational (non-experimental) studies, investigators observe individuals without experimental manipulation or intervention. There is an inadequacy about the term "observational study" because the outcome variable of an experiment could also be observed. Observational studies can be further categorized into descriptive and ...
Read More: Experimental Research Examples 2. Observational Study. Observational research is a non-experimental research method in which the researcher merely observes the subjects and notes behaviors or responses that occur (Ary et al., 2018).. This approach is unintrusive in that there is no manipulation or control exerted by the researcher.For instance, a researcher could study the ...
An example of an experimental study is a study where researchers had 20 test subjects and 10 were randomly assigned to receive a medicine being studied and the other 10 would receive a placebo ...
We may approach this study by 2 longitudinal designs: Prospective: we follow the individuals in the future to know who will develop the disease. Retrospective: we look to the past to know who developed the disease (e.g. using medical records) This design is the strongest among the observational studies. For example - to find out the relative ...
Experimental studies are ones where researchers introduce an intervention and study the effects. Experimental studies are usually randomized, meaning the subjects are grouped by chance. Randomized controlled trial (RCT): Eligible people are randomly assigned to one of two or more groups. One group receives the intervention (such as a new drug ...
Go to: Evidence-based medicine classifies studies into grades of evidence based on research architecture. 1, 2 This hierarchical approach to study design has been promoted widely in individual reports, meta-analyses, consensus statements, and educational materials for clinicians. For example, a prominent publication 3 reserved the highest grade ...
This kind of observational research can result in substantial profits. 5. Spying on Farms. Similar to the example above, observational research can also be implemented to study agriculture and farming. By using infrared imaging software from satellites, some companies can observe crops across the globe.
Diagnostic studies are classified as observational studies, but are a unique category and will be discussed independently. Interventional studies, also called experimental studies, are those where the researcher intercedes as part of the study design. Additionally, study designs may be classified by the role that time plays in the data ...
Cohort studies and case studies are considered observational in design, whereas the randomized controlled trial would be an experimental study. Let's take a closer look at the different types of observational study design. The 3 types of Observational Studies. The different types of observational studies are used for different reasons.
The key difference between observational studies and experimental designs is that a well-done observational study does not influence the responses of participants, ... Methods are the specific tools and procedures you use to collect and analyze data (for example, experiments, surveys, and statistical tests).
Here are the main advantages to expect for using experimental study vs observational experiment. Most experimental studies are shorter and smaller compared to observational studies. ... The research is an example observational study because it did not have any control. The researcher only observed the levels but did not have any type of control ...
Gain even more knowledge about experimental and observational studies thanks to the lesson called Experiments vs. Observational Studies: Definition, Differences & Examples. You will have the ...
The stripping of carbon nanotube forests (CNTF) is an urgent problem in terms of its application in thermal management and semiconductor devices. The easy‑peel and vertically aligned CNTF is prepared in a two‑step process of high vacuum magnetron sputtering and chemical vapor deposition. Based on the thick alumina buffer layer, CNTF with heights of 100 μm, 300 μm, and 500 μm was ...
This study involved the conduct of five experiments. The first experiment's goal was to assess each IOS parameter's capacity to identify SAD in patients with PPF. The study's criteria for diagnosing SAD were two out of the three small airway measurements (FEF25-75%, FEF50%, and FEF75%) having a predictive value of less than 65% according ...
The following analysis compares the resulting OFRFs to study the feasibility and stability of the hybrid method. Similar to the simulation examples, dynamic attributes are modified with a new constraint to represent the operating condition, making it easy and reliable to obtain the exact "operational" FRFs with the H 1 estimator.
This work presents an experimental realization of a ring scheme of nonlinear threshold controlled unidirectionally coupled (N = 3) second-order autonomous type oscillating systems. The originality of this work lies in having the threshold controller as the nonlinear element of the dynamical system and as the coupling element to form a ring circuit using these systems. The advantage of this ...