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 2,751–2,800 of 5,065 guides

5,065 guides
PART 04 · SUBPART 37 · GUIDE 433

Avoiding Spectrum Bias in Diagnostic Studies

How Do You Avoid Spectrum Bias in Diagnostic Research?

Spectrum bias can make a diagnostic test appear more accurate when researchers evaluate obvious cases and healthy controls rather than the patients who would actually receive the test. Learn how disease severity, competing diagnoses, recruitment, and clinical setting affect accuracy estimates and how to design more applicable studies.

PART 04 · SUBPART 37 · GUIDE 432

Index Test and Reference Standard Verification

Should Everyone Receive Both the Index Test and Reference Standard?

Applying the index test and reference standard to all eligible participants generally strengthens diagnostic accuracy research, but universal verification is not always ethical or feasible. Learn when incomplete verification is defensible, how it can distort accuracy estimates, and which design strategies can reduce bias.

PART 04 · SUBPART 37 · GUIDE 431

Reference Standards in Diagnostic Accuracy

What Is a Reference Standard, and What if No Perfect Standard Exists?

A reference standard establishes the target condition against which a diagnostic test is evaluated. Learn how to select a defensible standard, what alternatives exist when no perfect method is available, and how reference-standard errors affect sensitivity and specificity.

PART 04 · SUBPART 37 · GUIDE 430

Choosing an Index Test Comparator

What Should the Index Test Be Compared Against in a Diagnostic Accuracy Study?

An index test should ordinarily be evaluated against an appropriate reference standard, not simply another convenient test. Learn how to distinguish reference standards from competing diagnostic tests, select a credible comparison, and interpret accuracy when the reference standard is imperfect.

PART 04 · SUBPART 37 · GUIDE 429

When to Use a Diagnostic Accuracy Study

When Is a Diagnostic Accuracy Study the Right Design?

A diagnostic accuracy study evaluates how well a test identifies a target condition compared with an appropriate reference standard. Learn when this design is suitable, how to recruit representative participants, and why diagnostic accuracy alone does not establish clinical usefulness.

PART 04 · SUBPART 36 · GUIDE 428

When to Avoid Noninferiority Studies

When Should You Avoid a Noninferiority Study Design?

Noninferiority studies are not appropriate simply because demonstrating superiority would be difficult. Learn when uncertain comparator effectiveness, unjustified margins, weak assay sensitivity, or unacceptable treatment risks make another design more defensible.

PART 04 · SUBPART 36 · GUIDE 427

No Significant Difference vs. Equivalence

When Is “No Significant Difference” Not Evidence of Equivalence?

A nonsignificant difference does not establish that two treatments, methods, or groups are equivalent. Understand why conventional hypothesis testing cannot prove acceptable similarity, how equivalence tests work, and what researchers should conclude when results are inconclusive.

PART 04 · SUBPART 36 · GUIDE 425

Poor Adherence and Noninferiority Bias

Why Can Poor Adherence Bias a Noninferiority Trial Toward Success?

Poor adherence can make treatments appear more similar than they really are, potentially favoring a noninferiority conclusion. Understand when this occurs, why the direction of bias is not guaranteed, and how researchers can prevent and investigate adherence-related threats to validity.

PART 04 · SUBPART 36 · GUIDE 424

Assay Sensitivity in Noninferiority Trials

Why Does Assay Sensitivity Matter in Noninferiority Trials?

Assay sensitivity determines whether a noninferiority trial can distinguish an effective treatment from a less effective or ineffective one. Learn why this capability matters, how researchers establish its credibility, and why similar outcomes between treatment groups may otherwise be misleading.

PART 04 · SUBPART 36 · GUIDE 423

Risks of a Poorly Chosen Noninferiority Margin

Can a Poorly Chosen Noninferiority Margin Make an Inferior Treatment Look Acceptable?

A noninferiority trial can meet its statistical criterion even when the new treatment is meaningfully less effective. Understand how an overly permissive margin creates this risk, why statistical success does not validate the margin, and how researchers can recognize and prevent misleading conclusions.

PART 04 · SUBPART 36 · GUIDE 422

Noninferiority Margin and Its Justification

What Is a Noninferiority Margin, and How Should It Be Justified?

The noninferiority margin defines how much effectiveness a new intervention may lose before it becomes unacceptable. Learn how margins are established using historical evidence, effect preservation, and substantive judgment, and why an unjustified margin can undermine an otherwise well-conducted trial.

PART 04 · SUBPART 36 · GUIDE 421

Superiority vs. Noninferiority vs. Equivalence

Superiority vs. Noninferiority vs. Equivalence: Which Question Are You Actually Asking?

Superiority, noninferiority, and equivalence studies answer different research questions, even when they compare the same interventions. Understand how to select the appropriate objective, interpret confidence intervals, and avoid conclusions that the study was never designed to support.

PART 04 · SUBPART 36 · GUIDE 420

When to Use an Equivalence Study

When Is an Equivalence Study Appropriate?

An equivalence study tests whether differences between interventions or methods are sufficiently small to be considered acceptable in either direction. Understand when equivalence is the appropriate objective, how to justify its boundaries, and how to interpret the evidence.

PART 04 · SUBPART 36 · GUIDE 419

When to Use a Noninferiority Study

When Is a Noninferiority Study Appropriate?

A noninferiority study examines whether a new intervention is not unacceptably worse than an established comparator. Learn when this approach is justified, what makes its conclusions credible, and when another design would be more appropriate.

PART 04 · SUBPART 35 · GUIDE 418

Positioning Trials on the Pragmatic-Explanatory Continuum

How Do You Position a Trial on the Pragmatic-Explanatory Continuum?

Trials are rarely entirely explanatory or pragmatic. Learn how the PRECIS-2 framework evaluates nine design domains, how to interpret its five-point ratings, and how to position a trial according to the decisions its findings are intended to inform.

PART 04 · SUBPART 35 · GUIDE 417

Explanatory Trials and Real-World Applicability

Can Making a Trial More Explanatory Reduce Its Real-World Applicability?

Explanatory trials can clarify intervention effects under carefully specified conditions, but their findings may not transfer directly to routine practice. Learn which design choices can limit applicability, when those choices are justified, and how to distinguish credible efficacy evidence from claims about real-world effectiveness.

PART 04 · SUBPART 35 · GUIDE 416

Pragmatic Trials and Explaining Intervention Effects

Can Making a Trial More Pragmatic Make It Harder to Explain Why an Intervention Worked?

Pragmatic trials can establish whether introducing an intervention improves outcomes under ordinary conditions, but identifying why it worked may be more difficult. Learn how implementation variation, context, and multiple causal pathways complicate explanation and how carefully designed process evaluation can help.

PART 04 · SUBPART 35 · GUIDE 415

Choosing Meaningful Outcomes in Pragmatic Trials

Should Pragmatic Trials Prioritize Clinically Relevant Outcomes Over Convenient Measures?

A conveniently measured outcome is not necessarily the outcome that matters most to patients, practitioners, or policymakers. Learn how pragmatic trials should prioritize decision-relevant outcomes without overlooking measurement validity, feasibility, and the appropriate role of surrogate endpoints.

PART 04 · SUBPART 35 · GUIDE 414

Randomization and Blinding in Pragmatic Trials

Can a Pragmatic Trial Still Be Randomized and Blinded?

Pragmatic trials can be randomized and blinded, even when interventions are delivered under ordinary practice conditions. Learn which forms of masking remain feasible, how allocation concealment differs from blinding, and how to protect causal inference when participants or providers know their assignments.

PART 04 · SUBPART 35 · GUIDE 413

Intervention Standardization in Pragmatic Trials

How Much Intervention Standardization Is Compatible With a Pragmatic Trial?

Pragmatic trials do not require unrestricted intervention delivery. Learn how to standardize essential treatment components while allowing appropriate variation in providers, settings, and implementation procedures.

PART 04 · SUBPART 35 · GUIDE 412

Usual Care Variation in Pragmatic Trials

Should Usual Care Be Allowed to Vary Across Sites in a Pragmatic Trial?

Usual care rarely looks identical across institutions. Learn when pragmatic trials should preserve existing services, when differences threaten the intended comparison, and how to define, monitor, and analyze variation without creating an artificial control condition.

PART 04 · SUBPART 35 · GUIDE 411

Eligibility Criteria in Pragmatic Trials

How Broad Should Eligibility Criteria Be in a Pragmatic Trial?

Pragmatic trials often use broader eligibility criteria than explanatory trials, but including everyone is not necessarily appropriate. Learn how to define a target population, justify exclusions, and balance applicability, participant safety, and credible treatment comparisons.

PART 04 · SUBPART 35 · GUIDE 410

Intervention Efficacy vs. Effectiveness Trials

Is Your Trial Testing Whether an Intervention Can Work Under Ideal Conditions or Does Work in Usual Practice?

An intervention may produce benefits under carefully supported conditions without achieving the same results in routine practice. Learn how efficacy and effectiveness questions differ, why implementation conditions matter, and how to design a trial that estimates the effect relevant to your research or policy decision.

PART 04 · SUBPART 35 · GUIDE 409

Explanatory vs. Pragmatic Trials

Explanatory vs. Pragmatic Trial: What’s the Difference?

Explanatory and pragmatic trials differ primarily in the decisions they are designed to inform, not simply in how controlled or realistic they appear. Learn how their objectives influence eligibility, intervention delivery, comparison conditions, outcomes, and analysis, and why most trials occupy positions along a continuum.

PART 04 · SUBPART 34 · GUIDE 408

Intervention Failure vs. Implementation Failure

How Do You Distinguish Intervention Failure From Implementation Failure?

When an intervention produces disappointing results, the problem may lie in the intervention itself, how it was implemented, or the evaluation. Learn how fidelity and process evidence help distinguish these possibilities without treating poor implementation as an automatic explanation for null findings.

PART 04 · SUBPART 34 · GUIDE 407

Measuring Intervention Provider Competence

Should Intervention Provider Competence Be Measured?

Completing every prescribed intervention activity does not necessarily mean delivering it competently. Learn when provider competence should be assessed, how to distinguish competence from adherence, and how to evaluate delivery quality without confusing professional credentials with demonstrated performance.

PART 04 · SUBPART 34 · GUIDE 406

Participant Adherence and Intervention Fidelity

Is Participant Adherence Part of Intervention Fidelity?

An intervention may be delivered exactly as planned while participants engage with it inconsistently. Learn how participant adherence relates to intervention fidelity, why different frameworks define their relationship differently, and how to assess both without confusing delivery problems with participant behavior.

PART 04 · SUBPART 34 · GUIDE 405

Measuring Intervention Dose

Should Intervention Dose Be Measured in Research Studies?

Intervention dose describes how much treatment is provided or received, but attendance and duration do not always capture meaningful exposure. Learn when dose should be measured, which indicators are appropriate, and how dose information can improve interpretation without creating misleading causal claims.

PART 04 · SUBPART 34 · GUIDE 404

Intervention Flexibility and Treatment Fidelity

Can Excessive Intervention Flexibility Make It Difficult to Determine What Was Actually Tested?

Flexibility can make interventions responsive to participants and settings, but too much variation may leave researchers uncertain about what was actually delivered. Learn when adaptation threatens treatment fidelity, how to recognize intervention drift, and how to preserve an identifiable intervention without requiring identical delivery.

PART 04 · SUBPART 34 · GUIDE 403

Intervention Standardization and Real-World Applicability

Can Excessive Intervention Standardization Reduce Real-World Applicability?

Highly standardized interventions can improve delivery consistency, but they may also create conditions that are difficult to reproduce in ordinary practice. Learn when standardization limits applicability and how to preserve essential intervention functions while accommodating real-world variation.

PART 04 · SUBPART 34 · GUIDE 402

How Much Intervention Adaptation Is Appropriate?

How Much Adaptation Should Intervention Providers Be Allowed?

Intervention providers often need to adapt procedures to participant needs and local conditions. Learn how to establish defensible adaptation boundaries, distinguish faithful tailoring from consequential deviations, and evaluate whether modifications preserve the intervention being tested.

PART 04 · SUBPART 34 · GUIDE 401

Standardizing Intervention Components

Should Every Component of a Research Intervention Be Standardized?

Not every intervention component requires identical delivery. Learn how to distinguish essential functions from adaptable procedures, decide what must remain consistent across providers and sites, and design intervention standards that protect interpretability without eliminating justified flexibility.

PART 04 · SUBPART 34 · GUIDE 400

Assessing Intervention Delivery Fidelity

How Do You Assess Whether an Intervention Was Delivered as Intended?

A completed session does not necessarily mean an intervention was delivered as intended. Learn how to assess intervention delivery through structured observations, component checklists, dose records, and quality ratings, while distinguishing faithful adaptation from implementation deviations.

PART 04 · SUBPART 34 · GUIDE 399

Intervention Fidelity in Research Design

What Is Intervention Fidelity, and Why Should It Be Planned Before a Study Begins?

Intervention fidelity concerns whether a research intervention is implemented as intended, but fidelity involves more than following a protocol exactly. Learn how to define essential components, distinguish fidelity from adherence and adaptation, and build a defensible fidelity plan before data collection begins.

PART 04 · SUBPART 33 · GUIDE 398

Designing Studies With High Contamination Risk

How Should You Design a Study When Contamination Between Groups Is Likely?

When participants are likely to share intervention materials or influence one another, contamination should shape the research design from the beginning. Learn how to select an appropriate randomization unit, preserve meaningful treatment contrasts, and plan for unavoidable spillover.

PART 04 · SUBPART 33 · GUIDE 397

Measuring Contamination Between Study Groups

How Do You Measure Contamination Between Study Groups?

Measuring contamination requires more than counting control participants who encounter intervention materials. Learn how to define unintended exposure, assess its timing and intensity, calculate interpretable indicators, and evaluate the consequences for a treatment comparison.

PART 04 · SUBPART 33 · GUIDE 396

When Treatment Standardization Is Unrealistic

When Is Standardizing Treatments Across Study Groups Unrealistic in Real-World Research?

Real-world participants rarely receive identical services, and forcing uniform treatment can change the intervention being evaluated. Learn when standardization becomes impractical, how to distinguish meaningful variation from protocol deviations, and how to design interpretable studies without imposing artificial conditions.

PART 04 · SUBPART 33 · GUIDE 395

Standardizing Treatments Across Study Groups

Should Other Treatments or Services Be Standardized Across Study Groups?

Should intervention and control groups receive identical additional treatments and services? Learn when standardization protects a treatment comparison, when it changes the research question, and how to distinguish necessary controls from legitimate differences in care.

PART 04 · SUBPART 33 · GUIDE 394

Co-Interventions in Research

What Is Co-Intervention, and How Can It Distort Comparisons Between Study Groups?

Participants often receive treatments, services, or support beyond the intervention being studied. These co-interventions may influence outcomes, but they do not always constitute bias. Learn how to distinguish legitimate additional care from problematic differences and how to manage co-interventions without changing the research question.

PART 04 · SUBPART 33 · GUIDE 393

Staff-Related Intervention Contamination

How Can Staff Accidentally Deliver an Intervention to the Wrong Study Group?

Contamination does not always result from participants sharing materials. Researchers, instructors, clinicians, and intervention providers may inadvertently deliver experimental components to the wrong group. Understand how these errors arise, how they affect treatment comparisons, and how to prevent them without compromising participant safety or research integrity.

PART 04 · SUBPART 33 · GUIDE 392

Intervention Material Sharing and Contamination

How Does Sharing Intervention Materials With Control Participants Cause Contamination?

Intervention materials can spread from experimental participants to control participants through shared files, digital access, peer discussions, and staff practices. Learn when this sharing constitutes contamination, how it can influence treatment estimates, and which safeguards are methodologically appropriate.

PART 04 · SUBPART 33 · GUIDE 391

Cluster Randomization to Reduce Contamination

When Should You Use Cluster Randomization to Reduce Contamination?

Cluster randomization may reduce contamination when participants share intervention materials, providers cannot separate treatment conditions, or interventions operate through intact groups. Learn when this design is justified, what it cannot prevent, and how to account for its methodological costs.

PART 04 · SUBPART 33 · GUIDE 390

How Much Group Contamination Is Acceptable?

How Much Contamination Between Study Groups Is Acceptable?

There is no universal percentage of contamination that makes a study acceptable or invalid. Learn how contamination affects intervention comparisons, why its consequences depend on the estimand and outcome, and how researchers can establish defensible thresholds and responses before data collection.

PART 04 · SUBPART 32 · GUIDE 388

Minimizing Bias Without Blinding

How Do You Design a Study to Minimize Bias When Blinding Is Impossible?

Blinding is not always feasible, particularly in educational, behavioral, and complex intervention studies. Learn how to design credible unblinded research using safeguards that address specific sources of bias without pretending that every intervention can be concealed.

PART 04 · SUBPART 32 · GUIDE 387

When Side Effects Reveal Treatment Assignment

What Happens When Treatment Side Effects Reveal Group Assignment?

Even when treatments look identical, recognizable side effects may allow participants or clinicians to infer which treatment was received. Learn how this form of unblinding can influence expectations, outcome reporting, adherence, and the interpretation of randomized trial findings.

PART 04 · SUBPART 32 · GUIDE 386

Testing Whether Participants Can Guess Their Assignment

Should Researchers Test Whether Participants Can Guess Their Treatment Assignment?

Asking participants to guess their treatment assignment may appear to be a straightforward test of blinding, but correct guesses can reflect treatment effects, side effects, expectations, or chance. Learn when blinding assessments are useful, how to interpret them, and why they cannot independently establish whether a trial is biased.

PART 04 · SUBPART 32 · GUIDE 385

When Blinding Fails During Research

Can Blinding Fail During a Study?

Blinding can fail when participants, intervention providers, outcome assessors, or analysts learn treatment assignments that were intended to remain concealed. Understand how unblinding occurs, when it threatens validity, and what researchers should do when masking is compromised.

PART 04 · SUBPART 32 · GUIDE 384

Blinding and Objective Outcomes

Does Blinding Matter When Study Outcomes Are Objectively Measured?

Objectively measured outcomes may be less vulnerable to assessor expectations, but they are not automatically free from bias. Learn when blinding remains valuable, how treatment knowledge can affect apparently objective measurements, and when alternative safeguards may be sufficient.