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 2,501–2,550 of 5,065 guides

5,065 guides
PART 04 · SUBPART 58 · GUIDE 683

Sample-Size Sensitivity Analysis

Should Sample-Size Sensitivity Analysis Be Part of Study Planning?

A sample-size calculation may change substantially when its underlying assumptions change. Learn when sensitivity analysis is warranted, how to evaluate uncertain parameters, and how to document a sample-size decision that remains defensible under realistic alternatives.

PART 04 · SUBPART 58 · GUIDE 682

Using Plausible Assumptions for Sample Size

Can You Use a Range of Plausible Assumptions for Sample-Size Planning?

Sample-size calculations often depend on parameters that researchers cannot know precisely before collecting data. Learn how to develop evidence-informed ranges, compare alternative planning scenarios, and select a sample size that remains useful under uncertainty.

PART 04 · SUBPART 58 · GUIDE 681

Unknown Outcome Variability in Sample-Size Planning

What Should You Do When Outcome Variability Is Unknown for Sample-Size Planning?

When outcome variability is unknown, sample-size planning requires a defensible estimate or range rather than an arbitrary standard deviation. Learn how to use previous studies, pilot data, measurement characteristics, and sensitivity analysis to plan an informative study.

PART 04 · SUBPART 58 · GUIDE 680

Unknown Prevalence in Sample-Size Planning

What Should You Do When the Expected Prevalence Is Unknown for Sample-Size Planning?

When expected prevalence is unknown, using 50% can provide a conservative starting point for estimating a population proportion, but it is not always the most appropriate assumption. Learn how to evaluate available evidence, calculate sample size, and account for uncertainty and survey design.

PART 04 · SUBPART 58 · GUIDE 679

Why Pilot Study Effect Sizes Are Unstable

Why Are Pilot Study Effect Size Estimates Often Unstable?

Pilot studies can produce surprisingly large or small effect sizes even when the underlying population effect is unchanged. Understand the statistical reasons for this instability, how it can distort subsequent power calculations, and what researchers should do instead.

PART 04 · SUBPART 58 · GUIDE 678

Pilot Study Effect Sizes for Sample-Size Planning

Can a Pilot Study Provide a Reliable Effect Size for Sample-Size Planning?

Pilot studies can inform sample-size planning, but their observed effect sizes are often too uncertain to use directly. Learn when pilot estimates may be useful, what information is more reliable, and how to avoid planning an inadequately powered main study.

PART 04 · SUBPART 58 · GUIDE 676

Smallest Effect Size of Interest vs. Expected Effect Size

Should You Use the Smallest Effect Size of Interest Instead of the Expected Effect Size?

Should you calculate sample size using the effect you expect to observe or the smallest effect worth detecting? Understand how these assumptions serve different purposes and when a smallest effect size of interest provides a more defensible planning target.

PART 04 · SUBPART 58 · GUIDE 674

When the Expected Effect Size Is Unknown

What Should You Do When the Expected Effect Size Is Unknown?

An unknown expected effect size does not prevent defensible sample-size planning. Learn how to use meaningful effect thresholds, plausible assumptions, and sensitivity analyses without inventing an effect estimate.

PART 04 · SUBPART 57 · GUIDE 673

Revising a Study with an Unrealistic Sample Size

Should You Revise the Research Question, Design, Outcome, or Effect Size Assumption When the Required Sample Size Is Unrealistic?

When the required sample size is unrealistic, several study elements may be reconsidered, but they are not equally interchangeable. Learn when to revise the research question, improve the design, reconsider outcome measurement, or reassess effect size assumptions, and how to distinguish scientifically justified revisions from changes made merely to produce a smaller calculation.

PART 04 · SUBPART 57 · GUIDE 672

When the Required Sample Size Is Unattainable

What Should You Do When the Calculated Sample Size Is Impossible to Recruit?

A statistically justified sample size may exceed the number of participants a researcher can realistically recruit. Learn how to evaluate whether the calculation is appropriate, distinguish recruitment limitations from methodological problems, improve study efficiency, and decide when a smaller exploratory or feasibility study is more defensible than an inadequately powered confirmatory investigation.

PART 04 · SUBPART 57 · GUIDE 671

Different Outcomes and Sample Size Requirements

What Should You Do When Different Study Outcomes Require Very Different Sample Sizes?

Different study outcomes can produce substantially different sample size requirements, especially when they differ in variability, prevalence, effect magnitude, or statistical objectives. Learn how to evaluate competing calculations, distinguish essential outcomes from exploratory analyses, and decide whether to increase recruitment, revise the design, or narrow the study's confirmatory claims.

PART 04 · SUBPART 57 · GUIDE 670

Sample Size and the Primary Outcome

Should Sample Size Be Based Only on the Primary Outcome?

The primary outcome usually determines sample size in confirmatory research, but this does not mean that secondary outcomes are irrelevant or that one calculation is sufficient for every study. Learn when a primary-outcome calculation is appropriate, how co-primary outcomes change planning, and how to avoid making unsupported claims about analyses the study was not designed to power.

PART 04 · SUBPART 57 · GUIDE 669

Adjusting Sample Size for Nonresponse

How Do You Adjust Sample Size for Nonresponse?

A calculated survey sample size usually refers to the number of usable observations, not the number of people who must be invited. Learn how to adjust recruitment targets for expected nonresponse, distinguish response rates from participation yield, plan for subgroup differences, and understand why obtaining enough responses does not automatically eliminate nonresponse bias.

PART 04 · SUBPART 57 · GUIDE 668

Adjusting Sample Size for Dropout and Attrition

How Do You Adjust Sample Size for Expected Dropout or Attrition?

Expected dropout can reduce the number of participants contributing usable outcome data and compromise the precision or statistical power of a study. Learn how to calculate an attrition-adjusted recruitment target, distinguish dropout from missing measurements, plan for unequal retention across groups, and recognize why recruiting additional participants cannot automatically correct attrition bias.

PART 04 · SUBPART 57 · GUIDE 667

Design Effect and Sample Size Calculation

What Is the Design Effect, and When Do You Need It?

The design effect describes how a sampling design changes an estimator's variance relative to a reference design, often simple random sampling. Learn how to calculate and interpret design effects, when they increase or decrease sample size requirements, and why clustering, unequal weights, and stratification cannot always be represented by one universal adjustment.

PART 04 · SUBPART 57 · GUIDE 666

Clustering and Sample Size

How Does Clustering Affect Sample Size?

Clustering can substantially increase sample size requirements because participants within the same school, classroom, institution, or other group may provide correlated information. Learn how intraclass correlation, cluster size, the number of clusters, and the planned analysis influence statistical power and sample size, including why recruiting more participants from the same few clusters may be less useful than recruiting additional clusters.

PART 04 · SUBPART 57 · GUIDE 665

Repeated Measures and Sample Size

How Do Repeated Measures Affect Sample Size Requirements?

Repeated measurements can improve statistical efficiency, but collecting more observations from the same participants does not necessarily replace recruiting more people. Learn how within-person correlation, measurement timing, outcome variability, longitudinal models, and attrition influence sample size requirements.

PART 04 · SUBPART 57 · GUIDE 664

Unequal Allocation Ratio and Sample Size

How Does an Unequal Allocation Ratio Affect Sample Size?

Unequal allocation may be necessary when intervention costs, recruitment conditions, or ethical considerations make equal group sizes impractical. Learn how unequal allocation affects statistical power and total sample size, why equal allocation is often efficient, and when deliberately assigning more participants to one group can be methodologically justified.

PART 04 · SUBPART 57 · GUIDE 663

Number of Groups and Sample Size

How Does the Number of Groups Affect Sample Size?

Adding comparison groups can change the sample size needed for a study, but the relationship is not simply proportional. Learn how the number of groups interacts with the primary hypothesis, omnibus tests, pairwise comparisons, statistical power, multiple testing, and allocation decisions.

PART 04 · SUBPART 57 · GUIDE 662

Outcome Prevalence and Sample Size

How Does Outcome Prevalence Affect Sample Size?

Outcome prevalence affects the information available from binary outcomes and can substantially influence sample size requirements. Learn why 50% prevalence is commonly used for conservative proportion estimation, why rare outcomes can require large samples, and how prevalence interacts with precision, statistical power, and regression modeling.

PART 04 · SUBPART 57 · GUIDE 661

Outcome Variability and Sample Size

How Does Outcome Variability Affect Sample Size?

Outcome variability determines how precisely a study can estimate differences and how readily it can detect meaningful effects. Learn why larger standard deviations generally require larger samples, how variance enters common sample size formulas, and how to choose defensible variability assumptions using previous research, pilot evidence, and sensitivity analysis.

PART 04 · SUBPART 57 · GUIDE 660

One-Sided vs. Two-Sided Tests and Sample Size

One-Sided vs. Two-Sided Tests: How Do They Affect Sample Size?

One-sided tests may require fewer participants than two-sided tests when researchers test an effect in a prespecified direction. Learn why the difference occurs, how it changes statistical power and sample size, and why choosing a one-sided test merely to reduce recruitment requirements can undermine the research question.

PART 04 · SUBPART 57 · GUIDE 659

Significance Level and Sample Size

What Is Significance Level, and Why Does Alpha Affect Sample Size?

The significance level determines the threshold for rejecting a null hypothesis and influences the sample size needed to detect a specified effect. Learn what alpha = 0.05 actually means, why stricter thresholds generally require larger samples, and how to select an appropriate significance level without confusing statistical significance with practical importance.

PART 04 · SUBPART 57 · GUIDE 658

Effect Size and Sample Size Calculation

What Is Effect Size, and Why Does the Expected Effect Affect Sample Size?

Effect size describes the magnitude of a difference or relationship, and its assumed value can substantially change the sample size required for a study. Learn how raw and standardized effects differ, why small effects are harder to detect, and how to select meaningful effect assumptions without manipulating the calculation to fit available resources.

PART 04 · SUBPART 57 · GUIDE 657

Statistical Power in Research Design

What Is Statistical Power, and Why Does It Matter for Study Design?

Statistical power describes how likely a planned hypothesis test is to detect a specified effect under particular assumptions. Learn what 80% and 90% power mean, how effect size and sample size influence power, why underpowered studies can produce misleading interpretations, and how to use power analysis responsibly when designing research.

PART 04 · SUBPART 57 · GUIDE 656

Do All Quantitative Studies Need Sample Size Calculations?

Does Every Quantitative Study Need a Formal Sample Size Calculation?

Not every quantitative study requires a conventional statistical power calculation, but every study needs a defensible explanation of its sample size. Learn when formal calculations are essential, when alternative approaches may be appropriate, and how to distinguish statistical adequacy from recruitment feasibility.

PART 04 · SUBPART 57 · GUIDE 655

Factors Determining Research Sample Size

What Determines the Sample Size a Study Actually Needs?

The sample size a study needs depends on what it aims to estimate, detect, compare, or understand. Learn how research objectives, effect sizes, outcome variability, statistical power, sampling design, missing data, and analytical complexity influence sample size requirements, and why no universal participant count is sufficient for every study.

PART 04 · SUBPART 57 · GUIDE 654

Estimating Sample Size Without a Formal Calculation

How Do You Estimate Sample Size When No Formal Calculation Is Available?

A formal sample size calculation is not always available or meaningful, particularly for exploratory studies, novel outcomes, and research with limited preliminary information. Learn how to develop a defensible sample size estimate using the study's objectives, plausible assumptions, precision considerations, pilot evidence, and feasibility constraints without presenting arbitrary numbers as statistically justified.

PART 04 · SUBPART 57 · GUIDE 653

Sample Size and Statistical Power

How Many Participants Do You Need for Adequate Statistical Power?

There is no universal number of participants that makes a quantitative study adequately powered. Learn what determines sample size, how statistical power fits into the calculation, and why the planned analysis should come before the final number.

PART 04 · SUBPART 56 · GUIDE 652

Adapting Qualitative Sampling During Analysis

When Should Qualitative Sampling Decisions Change as Analysis Develops?

Qualitative sampling may need to evolve as analysis reveals missing perspectives, emerging categories, or unexplained contradictions. Learn when recruitment changes are methodologically justified, how to distinguish analytical adaptation from convenience, and how to document decisions without compromising research integrity.

PART 04 · SUBPART 56 · GUIDE 651

Excessive Homogeneity in Qualitative Sampling

Can a Qualitative Sample Be Too Homogeneous?

Homogeneous sampling can support focused qualitative inquiry, but excessive similarity may exclude experiences essential to understanding a phenomenon. Learn how to recognize when a sample is too narrow, distinguish justified focus from selection bias, and decide whether additional perspectives are necessary.

PART 04 · SUBPART 56 · GUIDE 650

Excessive Heterogeneity in Qualitative Sampling

Can a Qualitative Sample Be Too Heterogeneous?

A highly heterogeneous qualitative sample can reveal meaningful contrasts, but excessive variation may leave researchers unable to examine each experience or context adequately. Learn how to recognize when a sample is too broad, distinguish productive diversity from analytical fragmentation, and develop a sampling strategy that preserves sufficient depth.

PART 04 · SUBPART 56 · GUIDE 649

Negative and Deviant Case Sampling

Should Negative or Deviant Cases Be Deliberately Sampled in Qualitative Research?

Negative and deviant cases can reveal where emerging qualitative explanations fail, require qualification, or depend on overlooked conditions. Learn when to deliberately recruit them, how to distinguish genuine contradictions from unusual experiences, and how to integrate their evidence without overstating their significance.

PART 04 · SUBPART 56 · GUIDE 648

Identifying Essential Qualitative Participant Perspectives

How Do You Identify Essential Participant Perspectives in Qualitative Research?

Qualitative research may require perspectives from people who experience, implement, influence, or are affected by a phenomenon. Learn how to identify which viewpoints are essential, distinguish participant roles from analytical perspectives, and justify a sample that supports the research question without attempting to include everyone.

PART 04 · SUBPART 56 · GUIDE 647

Qualitative Sampling: Diversity vs. Depth

Should Qualitative Sampling Prioritize Diversity or Depth?

Qualitative sampling does not always benefit from recruiting the most diverse participants or the largest number of cases. Learn when diversity strengthens contextual understanding, when depth requires a narrower focus, and how to develop a sampling strategy that supports meaningful qualitative analysis.

PART 04 · SUBPART 56 · GUIDE 646

Snowball Sampling in Qualitative Research

When Is Snowball Sampling Useful in Qualitative Research?

Snowball sampling helps qualitative researchers identify participants through existing social or professional connections, particularly when relevant experiences are difficult to locate. Learn how to use referrals strategically, protect confidentiality, and avoid allowing recruitment networks to determine the limits of the analysis.

PART 04 · SUBPART 56 · GUIDE 645

When to Use Theoretical Sampling

When Is Theoretical Sampling Necessary?

Theoretical sampling allows emerging analysis to determine which cases, participants, or data are needed next. Learn when it is necessary in grounded theory, how it differs from initial purposive sampling, and how to document decisions that contribute to theoretical development.

PART 04 · SUBPART 56 · GUIDE 644

When to Use Criterion Sampling

When Is Criterion Sampling Useful?

Criterion sampling deliberately selects cases that meet conditions central to a research question. Learn when it is appropriate, how to establish and verify criteria, and why meeting eligibility conditions does not automatically guarantee information richness or analytical adequacy.

PART 04 · SUBPART 56 · GUIDE 643

Maximum Variation vs. Homogeneous Sampling

Maximum Variation vs. Homogeneous Sampling: Which Fits Your Qualitative Research Question?

Maximum variation sampling examines a phenomenon across meaningfully different cases, while homogeneous sampling supports deeper investigation within a relatively specific group. Learn how to choose between them, define appropriate selection criteria, and avoid confusing qualitative diversity with statistical representativeness.

PART 04 · SUBPART 56 · GUIDE 642

Purpose of Purposive Sampling in Qualitative Research

What Is the Purpose of Purposive Sampling in Qualitative Research?

Purposive sampling helps qualitative researchers deliberately select cases relevant to understanding a phenomenon. Learn why it is used, how different purposive strategies support analytical objectives, and why its purpose is not to reproduce population proportions.

PART 04 · SUBPART 56 · GUIDE 641

Selecting Information-Rich Participants

How Do You Select Information-Rich Participants for Qualitative Research?

Information-rich participants can provide experiences and perspectives directly relevant to a qualitative research question. Learn how to identify them, establish defensible selection criteria, avoid relying on status or accessibility alone, and determine whether the resulting sample is analytically adequate.

PART 04 · SUBPART 56 · GUIDE 640

Sampling in Qualitative Research

How Do You Sample Participants in Qualitative Research?

Qualitative sampling is usually about selecting participants or cases that can illuminate the phenomenon, not reproducing a population in miniature. Learn how to build a sample that fits your qualitative methodology and analytical purpose.

PART 04 · SUBPART 55 · GUIDE 639

When Convenience Recruitment Bias Becomes Unacceptable

When Does Convenience-Based Recruitment Create Unacceptable Selection Bias?

Convenience recruitment becomes methodologically unacceptable when selection mechanisms threaten the study's central conclusions and cannot be addressed through defensible design or analytical assumptions. Learn how to recognize serious bias, evaluate its consequences, and decide when redesign is necessary.

PART 04 · SUBPART 55 · GUIDE 638

Avoiding Easily Accessible Participant Bias

How Do You Avoid Recruiting Only Easily Accessible Participants?

Recruiting only easily accessible participants can systematically exclude people whose experiences are important to a study. Learn how to identify recruitment blind spots, diversify access routes, and strengthen sampling decisions without making unsupported claims about representativeness.

PART 04 · SUBPART 55 · GUIDE 637

Can Quotas Make Convenience Samples Representative?

Can Quota Sampling Make a Convenience Sample Representative?

Quota sampling can make a convenience sample resemble the population on selected characteristics, but matching demographic proportions does not guarantee representativeness. Learn why within-category selection matters, how adjustment methods work, and what researchers can legitimately claim.

PART 04 · SUBPART 55 · GUIDE 636

Setting Sampling Quotas for Population Groups

When Should You Set Sampling Quotas for Important Population Groups?

Sampling quotas can prevent important population groups from being overlooked during non-probability recruitment. Learn when quotas are justified, how to determine appropriate category targets, and why satisfying quotas does not establish statistical representativeness.

PART 04 · SUBPART 55 · GUIDE 635

Reducing Non-Probability Sampling Limitations

How Can Research Design Reduce the Limitations of Non-Probability Sampling?

Non-probability sampling has important inferential limitations, but thoughtful research design can reduce avoidable weaknesses. Learn how to improve recruitment coverage, document selection, use auxiliary information, strengthen causal comparisons, and report conclusions that remain defensible.

PART 04 · SUBPART 55 · GUIDE 634

Non-Probability Sampling and Causal Inference

Can a Non-Probability Sample Support Causal Inference?

Non-probability sampling does not automatically prevent causal inference. A study may identify a credible causal effect among recruited participants while remaining unable to establish whether that effect applies to a wider population. Learn how treatment assignment, confounding, selection, and transportability determine what causal conclusions are defensible.