Manuel B. Garcia

Manuel B. Garcia serves as the Senior Director for Educational Technology and Digital Learning at FEU Institute of Technology, Manila, Philippines. Read More

Contact Info

1607, FEU Tech Building,
P. Paredes St, Sampaloc,
Manila, Philippines
mbgarcia@feutech.edu.ph

Follow Me

The Complete Library

All Research Guides

Browse every published guide in the Researcher's Field Guide, including the complete research journey and supplemental collections, listed alphabetically for quick reference.

Complete Library

Every published guide

Showing 2,251–2,300 of 5,065 guides

5,065 guides
PART 04 · SUBPART 80 · GUIDE 933

When Missing Data Make a Research Design Unacceptable

When Does Anticipated Missing Data Make a Research Design Unacceptable?

A research design may become unacceptable when predictable missingness prevents credible estimation of its primary outcome or effect. Learn how to distinguish manageable missing data from fundamental design failures, evaluate alternatives, and decide when the research question or measurement strategy must change.

PART 04 · SUBPART 80 · GUIDE 932

How Much Missing Data Is Too Much?

How Much Missing Data Would Make Your Primary Research Question Difficult to Answer?

There is no universal percentage of missing data that makes a study invalid. The consequences depend on which information is missing, why it is unavailable, and whether the remaining observations can support the primary analysis. Learn how to assess missingness using precision, identification assumptions, and sensitivity analyses rather than arbitrary cutoffs.

PART 04 · SUBPART 80 · GUIDE 931

Retaining Partial Data After Participant Withdrawal

Should Partial Data Be Retained When Participants Withdraw from Some Study Activities?

Participant withdrawal does not always require discarding every previously collected observation. Learn how to distinguish withdrawal from specific activities, intervention discontinuation, and withdrawal of consent, while protecting participant rights and preserving scientifically useful data where permitted.

PART 04 · SUBPART 80 · GUIDE 930

Documenting Reasons for Missing Research Data

Should Researchers Collect and Document Reasons for Missing Data?

Recording why research data are missing can improve study monitoring, reveal preventable collection problems, and inform sensitivity analyses. Learn what reasons to document, how to classify them consistently, and why recorded explanations do not automatically establish the statistical missing-data mechanism.

PART 04 · SUBPART 80 · GUIDE 929

Backup Data Sources for Critical Research Outcomes

Should Critical Research Outcomes Have Backup Data Sources?

Backup data sources can preserve critical outcomes when planned assessments are unavailable, but alternative records must measure the intended outcome accurately. Learn when backup sources are justified, how to establish source equivalence, and what to specify before data collection.

PART 04 · SUBPART 80 · GUIDE 928

Shorter Questionnaires and Missing Data

Can Shortening Research Questionnaires Reduce Missing Data?

Shortening a questionnaire may reduce respondent burden and prevent some missing data, but removing important items can weaken measurement validity. Learn how to identify unnecessary questions, evaluate survey breakoff, and improve completion without sacrificing the information your study needs.

PART 04 · SUBPART 80 · GUIDE 927

Changing Measurement Schedules to Reduce Missing Data

Can Changing the Measurement Schedule Reduce Missing Data?

Changing when and how often researchers collect measurements can reduce preventable missingness, but poorly chosen schedules may alter the outcomes being measured. Learn how to improve follow-up timing while preserving scientific validity and comparability.

PART 04 · SUBPART 80 · GUIDE 926

Missing Data Patterns Between Study Groups

Why Might Missing Data Patterns Differ Between Study Groups?

Missing data may occur differently across study groups because of participant characteristics, intervention demands, or outcome-related processes. Learn how to distinguish unequal missingness rates from biased comparisons and how to investigate and address group-specific missing-data risks.

PART 04 · SUBPART 80 · GUIDE 925

Identifying Variables at Risk of Missing Data

How Do You Identify Which Research Variables Are Most Likely to Have Missing Data?

Some research variables are more likely to be missing because of their sensitivity, timing, complexity, or collection procedures. Learn how to identify vulnerable measurements before data collection and prioritize practical prevention without confusing missingness risk with statistical importance.

PART 04 · SUBPART 79 · GUIDE 923

Explaining Confounding Control and Limitations

How Do You Explain Which Confounders Your Research Design Can and Cannot Address?

Explaining confounding control requires more than listing adjusted variables or acknowledging possible bias. Learn how to distinguish addressed and unresolved confounding, evaluate its consequences, and communicate the strength of your causal conclusions.

PART 04 · SUBPART 79 · GUIDE 922

Planning Confounding Control Before Data Collection

Should You Plan Your Confounding Control Strategy Before Data Collection?

Confounding control begins before statistical analysis. Learn how to develop a defensible strategy that connects your causal question, participant selection, confounder measurement, analytical methods, and sensitivity analyses before data collection starts.

PART 04 · SUBPART 79 · GUIDE 921

Mediator and Collider Adjustment Bias

Can Adjusting for Mediators or Colliders Introduce Bias?

Adjusting for a mediator can remove part of the causal effect you intend to estimate, while conditioning on a collider can create a misleading association. Understand the different mechanisms, practical examples, and how to avoid inappropriate adjustment.

PART 04 · SUBPART 79 · GUIDE 920

Excessive Measurement and Overadjustment

Can Measuring Too Many Variables Encourage Unnecessary Statistical Adjustment?

Collecting many variables does not itself cause overadjustment, but it can encourage poorly justified statistical models. Learn how excessive measurement, causal misclassification, and indiscriminate covariate adjustment can compromise research findings.

PART 04 · SUBPART 79 · GUIDE 919

Insufficient Confounder Measurement

Can Measuring Too Few Confounders Undermine Causal Inference?

Failing to measure important confounders can leave observational findings vulnerable to alternative causal explanations, even after statistical adjustment. Learn how to recognize insufficient measurement, assess its consequences, and strengthen research design before data collection.

PART 04 · SUBPART 79 · GUIDE 918

Measuring Variables for Possible Adjustment

Should You Measure a Variable Just Because You Might Adjust for It Later?

Collecting extra variables in case they become useful for statistical adjustment may seem cautious, but more data do not automatically produce more credible results. Learn how to decide which variables deserve measurement before a study begins.

PART 04 · SUBPART 79 · GUIDE 916

Matching Participants to Reduce Confounding

Should You Match Participants to Reduce Confounding?

Matching participants can improve comparability between study groups on selected characteristics, but it does not automatically eliminate confounding. Understand when matching is appropriate, which variables to match, and how to evaluate the resulting sample and analysis.

PART 04 · SUBPART 79 · GUIDE 915

Restricting Eligibility to Reduce Confounding

Should You Restrict Participant Eligibility to Reduce Confounding?

Restricting eligibility can reduce confounding by limiting variation in selected participant characteristics, but it also narrows the population studied. Understand when restriction is appropriate, when it may introduce problems, and how to justify eligibility criteria.

PART 04 · SUBPART 79 · GUIDE 914

Identifying Potential Confounders Before a Study

Which Potential Confounders Should You Identify Before Conducting a Study?

Identifying potential confounders before data collection requires more than listing variables associated with an outcome. Learn how to distinguish plausible common causes from mediators, colliders, and other variables that may not belong in an adjustment strategy.

PART 04 · SUBPART 79 · GUIDE 913

Effect Modifiers, Moderators, and Confounders

What Are Effect Modifiers and Moderators, and Why Aren’t They the Same as Confounders?

Confounders distort the relationship researchers want to estimate, while effect modifiers and moderators describe situations in which that relationship differs across levels of another variable. The distinction affects what you adjust for, what you report, and what your findings actually mean.

PART 04 · SUBPART 79 · GUIDE 912

Statistical Adjustment and Confounding

Why Doesn’t Statistical Adjustment Automatically Fix Confounding?

Statistical adjustment can reduce confounding when the right variables are measured and modeled appropriately, but it cannot guarantee an unbiased estimate. Unmeasured confounders, measurement error, model misspecification, and inappropriate adjustment can leave or even introduce bias.

PART 04 · SUBPART 79 · GUIDE 911

Confounding in Research

What Is Confounding, and Why Can It Distort a Relationship?

Confounding occurs when the relationship you are trying to estimate becomes mixed with the influence of another factor. Identifying a confounder requires causal reasoning, not simply finding a variable associated with the outcome.

PART 04 · SUBPART 78 · GUIDE 910

Designing Validation Checks for Research Variables

How Do You Design Validation Checks for Critical Research Variables?

A complete dataset is not necessarily an accurate one. Learn how to identify critical research variables, design meaningful validation checks, distinguish data errors from legitimate observations, and document corrections without compromising research integrity.

PART 04 · SUBPART 78 · GUIDE 909

Standardization vs. Blinding for Information Bias

When Is Standardization More Important Than Blinding for Preventing Information Bias?

Blinding reduces the influence of knowing study-group status, while standardization addresses differences in how measurements are collected. Learn when procedural consistency deserves priority, when assessor blinding is essential, and how to combine both safeguards without confusing their purposes.

PART 04 · SUBPART 78 · GUIDE 908

Blinding to Prevent Information Bias

When Is Blinding the Most Effective Protection Against Information Bias?

Blinding can reduce information bias when knowledge of treatment or exposure status influences measurement, reporting, or outcome assessment. Learn which research roles should be blinded, why subjective outcomes require particular attention, and when standardization or independent assessment may be more effective.

PART 04 · SUBPART 78 · GUIDE 907

Standardized Measurement Across Study Groups

Should Measurement Procedures Be Identical Across Study Groups?

Using identical measurement procedures across study groups can reduce information bias, but identical procedures do not always produce equivalent measurements. Learn when standardization is essential, when justified adaptations are appropriate, and how to establish comparability without sacrificing construct validity.

PART 04 · SUBPART 78 · GUIDE 906

Device Calibration and Information Bias

Can Differences in Device Calibration Introduce Information Bias?

Differences in device calibration can make identical underlying quantities produce different recorded measurements. Learn how systematic error, calibration drift, and device allocation can distort research comparisons, and how researchers can establish measurement comparability through calibration, verification, and sensitivity analysis.

PART 04 · SUBPART 78 · GUIDE 905

Medical Record Completeness and Information Bias

Can Differences in Medical Record Completeness Between Groups Introduce Information Bias?

Medical records do not capture every clinical event equally. Learn how differences in documentation, healthcare contact, and data availability between study groups can distort exposure and outcome measurement, and how researchers can evaluate record completeness before drawing conclusions.

PART 04 · SUBPART 78 · GUIDE 904

Outcome Assessor Bias and Exposure Knowledge

Can Outcome Assessor Bias Arise When Assessors Know Participants' Exposure Status?

Outcome assessors may interpret or score the same evidence differently when they know participants' treatment or exposure status. Learn when this creates detection bias, why subjective outcomes are especially vulnerable, and how blinding, standardized scoring, and independent assessment can strengthen research comparisons.

PART 04 · SUBPART 78 · GUIDE 903

Interviewer Bias Across Study Groups

Can Interviewer Bias Arise When Questions Are Asked Differently Across Study Groups?

Interviewers can influence research findings through differences in question wording, probing, tone, and interpretation across study groups. Learn when these differences create information bias, how they affect exposure and outcome measurement, and which design safeguards improve comparability.

PART 04 · SUBPART 78 · GUIDE 902

Recall Bias in Case-Control Studies

Can Recall Bias Cause Cases to Remember Past Exposures Differently from Controls?

Cases and controls may remember or report past exposures differently because of their outcome status, beliefs, or experiences. Learn how differential recall can distort case-control associations, why its direction is not predictable without assumptions, and how researchers can design more credible retrospective exposure assessments.

PART 04 · SUBPART 78 · GUIDE 901

Reporting Bias in Exposure Measurement

Can Participants' Exposure Reports Be Affected by Conscious or Unconscious Reporting Bias?

Participants may overreport, underreport, or misremember their exposure to behaviors, interventions, and environmental conditions. Learn how conscious and unconscious reporting processes affect exposure measurement, when errors distort associations, and how researchers can evaluate and reduce the problem.

PART 04 · SUBPART 77 · GUIDE 899

Selection Bias Risks Before Data Analysis

What Selection Bias Risks Should Researchers Address Before Planning Data Analysis?

Selection bias often begins before researchers open their statistical software. Learn how to audit sampling frames, eligibility rules, recruitment, consent, follow-up, and analytical inclusion before choosing adjustment methods, and distinguish problems that statistical analysis can address from those requiring better design.

PART 04 · SUBPART 77 · GUIDE 898

Matching, Confounding, and Selection Bias

Can Matching Reduce Confounding but Introduce Selection Bias?

Matching can improve comparability between exposed and unexposed groups, but selecting matches is not automatically harmless. Learn when matching reduces confounding, how inappropriate matching variables can introduce bias, and why excluding unmatched participants may change the causal effect being estimated.

PART 04 · SUBPART 77 · GUIDE 897

Restriction, Confounding, and Generalizability

Can Restriction Reduce Confounding but Limit Generalizability?

Restriction can reduce confounding by limiting variation in selected characteristics, but it also changes the population represented by a study. Learn when restriction improves causal comparability, how it affects generalizability and precision, and why restricting on the wrong variable may introduce bias.

PART 04 · SUBPART 77 · GUIDE 896

Designing Comparison Groups to Minimize Selection Bias

How Do You Design a Comparison Group to Minimize Selection Bias?

A comparison group must represent a credible alternative to the exposure or intervention being studied. Learn how to establish common eligibility, timing, and outcome measurement, distinguish confounding from selection bias, and avoid comparison groups that create misleading causal estimates.

PART 04 · SUBPART 77 · GUIDE 895

Conditioning on Participation and Selection Bias

Can Conditioning on Study Participation Create Misleading Associations?

Restricting analysis to people who participate in a study can create associations that do not exist in the intended population. Learn how collider bias arises through selection, why participation can distort otherwise independent variables, and how causal diagrams help researchers identify problematic conditioning decisions.

PART 04 · SUBPART 77 · GUIDE 894

Loss to Follow-Up and Selection Bias

Can Loss to Follow-Up Introduce Selection Bias?

Losing participants during follow-up can change the composition of a study and distort its findings. Learn when attrition creates selection bias, why equal dropout rates do not guarantee unbiased results, and how researchers can plan follow-up and analysis strategies that preserve credible inference.

PART 04 · SUBPART 77 · GUIDE 893

Informed Consent and Participant Selection

Can Informed Consent Requirements Affect Who Participates in a Study?

Informed consent protects participants' autonomy, but the way consent is presented and obtained can influence who enters a study. Learn how consent procedures may shape sample composition, when consent-related nonparticipation threatens inference, and how researchers can improve accessibility without weakening ethical protections.

PART 04 · SUBPART 77 · GUIDE 892

Recruitment Channels and Selection Bias

Can Recruitment Channels Introduce Selection Bias?

Recruitment channels determine who encounters a research invitation and who has an opportunity to participate. Learn how different channels can systematically exclude eligible people, when this threatens population estimates or causal comparisons, and how to evaluate recruitment coverage before data collection.

PART 04 · SUBPART 77 · GUIDE 891

Eligibility Criteria and Selection Bias

Can Eligibility Criteria Introduce Selection Bias?

Eligibility criteria help define a research population, but some restrictions can distort associations or causal comparisons. Learn when inclusion and exclusion rules create selection bias, when they primarily limit generalizability, and how to justify eligibility decisions before recruitment.

PART 04 · SUBPART 77 · GUIDE 890

Selection Bias Before Recruitment

How Can Selection Bias Arise Before Participant Recruitment Begins?

Selection bias can originate before researchers approach a single participant. Learn how decisions about study populations, sampling frames, research sites, and eligibility can shape who becomes observable, when selection threatens causal inference, and how to identify these risks during research design.

PART 04 · SUBPART 76 · GUIDE 889

Design Evidence for Construct Validity

What Design Evidence Is Needed to Establish That a Study Tested Its Intended Construct?

A study cannot establish that it tested its intended construct through statistical significance, instrument reliability, or intervention labels alone. Learn what theoretical, procedural, measurement, and analytical evidence is needed to support defensible construct claims.

PART 04 · SUBPART 76 · GUIDE 888

Triangulation and Construct Validity

Can Triangulation Strengthen Confidence That a Research Construct Was Adequately Captured?

Triangulation can strengthen confidence in a research construct by examining it through complementary sources, methods, or perspectives. Learn when convergence supports interpretation, why disagreement can be informative, and how to avoid treating triangulation as automatic validation.

PART 04 · SUBPART 76 · GUIDE 887

Using Multiple Measures for Construct Validity

Can Using Multiple Measures Strengthen Construct Validity?

Using multiple measures can strengthen construct validity when they provide complementary evidence about the intended construct. Learn how to select appropriate indicators, evaluate agreement and disagreement between measures, and avoid assuming that more instruments automatically produce better measurement.

PART 04 · SUBPART 76 · GUIDE 886

Narrow Operationalization and Construct Validity

Can Operationalizing a Complex Construct Too Narrowly Threaten Construct Validity?

Complex constructs such as engagement, critical thinking, and self-regulated learning cannot always be represented adequately by one convenient indicator. Learn when narrow operationalization threatens construct validity, how to distinguish focused measurement from underrepresentation, and what evidence supports a defensible interpretation.

PART 04 · SUBPART 76 · GUIDE 885

Differences in Survey Item Interpretation

Can Participants Interpret the Same Survey or Questionnaire Item Differently?

Identical questionnaire wording does not guarantee that respondents understand a question in the same way. Learn how differences in language, experience, context, and response processes can distort survey findings, and how to investigate item interpretation before data collection.

PART 04 · SUBPART 76 · GUIDE 884

Measuring a Related Rather Than Intended Construct

Can a Research Instrument Measure a Related Construct Instead of the Intended Construct?

An instrument can produce consistent scores and correlate with the intended construct while primarily measuring something else. Learn how to recognize construct substitution, distinguish related concepts, and collect evidence supporting the interpretation your research requires.