{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T05:46:16Z","timestamp":1790660776005,"version":"4.1.0"},"update-to":[{"DOI":"10.12688\/f1000research.185853.1","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000}}],"reference-count":49,"publisher":"F1000 Research Ltd","license":[{"start":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["f1000research.com"],"crossmark-restriction":false},"short-container-title":["F1000Res"],"abstract":"<ns3:p>Abstract Background Sample size determination is one of the least reported elements in the methods sections of health professions education (HPE) research. Underpowered studies miss real effects, whereas overpowered studies waste participants and resources. Guidance remains fragmented across statistical textbooks and discipline-specific papers, with no unified reference for HPE researchers. Methods We developed this guide as a structured narrative synthesis. Sources were identified through targeted database searching, citation chaining, and standard statistical texts, and selected against pre-specified criteria; where recommendations conflicted, precedence was given to the primary methodological source and to empirical evidence generated within HPE. Rather than proposing a universal formula, which does not exist, we organised sample-size logic by study design. We set out a stepwise workflow and emphasise that the expected difference should be anchored to previous evidence and to educational importance, with pilot work used to estimate variability, recruitment and attrition rather than the effect size itself. For each design we give the statistical rationale, the applicable formula or rule, a worked example with fully stated assumptions, and reporting expectations. Results The designs covered include two-group comparisons of means and proportions, pre\u2013post and repeated-measures designs, cluster-randomized educational trials, survey and questionnaire studies, Delphi and consensus studies, pilot and feasibility studies, qualitative enquiry, simulation-based education, and instrument validation. The guide concludes with a reporting checklist for the methods section, serving investigators, peer reviewers, and journal editors. Precision-based planning, software implementation, and designs beyond closed-form calculation are addressed separately. Conclusions Sample size in HPE research is a design-specific judgment, not a formula-specific calculation. The numerical benchmarks reproduced here are illustrative anchors rather than methodological minima. This guide consolidates that judgment into a practical reference reflecting the methodological diversity of the field.<\/ns3:p>","DOI":"10.12688\/f1000research.185853.2","type":"journal-article","created":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T08:54:06Z","timestamp":1788425646000},"page":"1147","update-policy":"https:\/\/doi.org\/10.12688\/f1000research.crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Sample size determination across study designs in health professions education research: a methodology guide"],"prefix":"10.12688","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4484-8225","authenticated-orcid":false,"given":"Waqar M","family":"Naqvi","sequence":"first","affiliation":[{"name":"Department of Health Professions Education, Faculty of Interdisciplinary Sciences Wardha Datta Meghe Institute of Higher Education and Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7024-4594","authenticated-orcid":false,"given":"Anusha","family":"Divvi","sequence":"additional","affiliation":[{"name":"Public Health Dentistry, Indira Gandhi Institute of Dental Sciences Puducherry Puducherry Sri Balaji Vidyapeeth University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6868-9639","authenticated-orcid":false,"given":"Shivashankar","family":"Kengadaran","sequence":"additional","affiliation":[{"name":"Public Health Dentistry, Indira Gandhi Institute of Dental Sciences Puducherry Puducherry Sri Balaji Vidyapeeth University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7414-2242","authenticated-orcid":false,"given":"Aishwarya A","family":"Pashine","sequence":"additional","affiliation":[{"name":"Department of Technical Medicine (Medical Simulation), Faculty of Interdisciplinary Sciences Wardha Datta Meghe Institute of Higher Education and Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maheshkumar","family":"Baladaniya","sequence":"additional","affiliation":[{"name":"Department of Physical Therapy New York Neighborhood Physical Therapy PC"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aditya","family":"Patel","sequence":"additional","affiliation":[{"name":"Department of Conservative Dentistry and Endodontics, Sharad Pawar Dental College and Hospital Wardha Datta Meghe Institute of Higher Education and Research"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2669-4488","authenticated-orcid":false,"given":"Praveen Kumar","family":"Kandakurthi","sequence":"additional","affiliation":[{"name":"College of Health Sciences Ajman Ajman Gulf Medical University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0839-5943","authenticated-orcid":false,"given":"Ramprasad","family":"Muthukrishnan","sequence":"additional","affiliation":[{"name":"College of Health Sciences Ajman Ajman Gulf Medical University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"2560","published-online":{"date-parts":[[2026,9,3]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"73-83","DOI":"10.1007\/s10459-014-9509-5","article-title":"Got power? A systematic review of sample size adequacy in health professions education research.","volume":"20","author":"D Cook","year":"2015","journal-title":"Adv Health Sci Educ Theory Pract."},{"key":"ref2","doi-asserted-by":"publisher","first-page":"365-376","DOI":"10.1038\/nrn3475","article-title":"Power failure: why small sample size undermines the reliability of neuroscience.","volume":"14","author":"K Button","year":"2013","journal-title":"Nat Rev Neurosci."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"227-238","DOI":"10.1111\/j.1365-2923.2010.03890.x","article-title":"Method and reporting quality in health professions education research: a systematic review.","volume":"45","author":"D Cook","year":"2011","journal-title":"Med Educ."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1002-1009","DOI":"10.1001\/jama.298.9.1002","article-title":"Association between funding and quality of published medical education research.","volume":"298","author":"D Reed","year":"2007","journal-title":"JAMA."},{"key":"ref5","doi-asserted-by":"publisher","first-page":"e5661","DOI":"10.1136\/bmj.e5661","article-title":"Consort 2010 statement: extension to cluster randomised trials.","volume":"345","author":"M Campbell","year":"2012","journal-title":"BMJ."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"19-24","DOI":"10.1198\/000313001300339897","article-title":"The abuse of power: the pervasive fallacy of power calculations for data analysis.","volume":"55","author":"J Hoenig","year":"2001","journal-title":"Am Stat."},{"key":"ref7","doi-asserted-by":"publisher","first-page":"1291","DOI":"10.12688\/f1000research.141173.3","article-title":"Exploring sample size determination in educational research: a comprehensive review.","volume":"12","author":"S Besekar","year":"2024","journal-title":"F1000Res."},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.4324\/9780203771587","author":"J Cohen","year":"2013","journal-title":"Statistical Power Analysis for the Behavioral Sciences."},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1201\/9781315183084","author":"S Chow","year":"2017","journal-title":"Sample Size Calculations in Clinical Research."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1002\/9781118874905","author":"D Machin","year":"2018","journal-title":"Sample sizes for clinical, laboratory and epidemiology studies."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"537-563","DOI":"10.1146\/annurev.psych.59.103006.093735","article-title":"Sample size planning for statistical power and accuracy in parameter estimation.","volume":"59","author":"S Maxwell","year":"2008","journal-title":"Annu Rev Psychol."},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1201\/b18676","article-title":"Power Analysis of Trials with Multilevel Data.","author":"M Moerbeek","year":"2015","journal-title":"Chapman and Hall\/CRC."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"Doc26","DOI":"10.3205\/zma001103","article-title":"Cluster-randomized Studies in Educational Research: Principles and Methodological Aspects.","volume":"34","author":"J Dreyhaupt","year":"2017","journal-title":"GMS J Med Educ."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"15-27","DOI":"10.21315\/mjms2021.28.2.2","article-title":"A Step-by-Step Process on Sample Size Determination for Medical Research.","volume":"28","author":"M Bujang","year":"2021","journal-title":"Malays J Med Sci."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"1292-1300","DOI":"10.1093\/ije\/dyl129","article-title":"Sample size for cluster randomized trials: effect of coefficient of variation of cluster size and analysis method.","volume":"35","author":"S Eldridge","year":"2006","journal-title":"Int J Epidemiol."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"1685-1704","DOI":"10.1002\/sim.4780111304","article-title":"Repeated measures in clinical trials: analysis using mean summary statistics and its implications for design.","volume":"11","author":"L Frison","year":"1992","journal-title":"Stat Med."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"287-291","DOI":"10.1002\/pst.185","article-title":"Sample size of 12 per group rule of thumb for a pilot study.","volume":"4","author":"S Julious","year":"2005","journal-title":"Pharm Stat."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"1753-1760","DOI":"10.1177\/1049732315617444","article-title":"Sample Size in Qualitative Interview Studies: Guided by Information Power.","volume":"26","author":"K Malterud","year":"2016","journal-title":"Qual Health Res."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"335-340","DOI":"10.3102\/10769986027004335","article-title":"Sample size requirements for testing and estimating coefficient alpha.","volume":"27","author":"D Bonett","year":"2002","journal-title":"J Educ Behav Stat."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"5-9","DOI":"10.1016\/j.ijnurstu.2014.09.006","article-title":"Basic statistical reporting for articles published in biomedical journals: the \"Statistical Analyses and Methods in the Published Literature\" or the SAMPL Guidelines.","volume":"52","author":"T Lang","year":"2015","journal-title":"Int J Nurs Stud."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"175-191","DOI":"10.3758\/bf03193146","article-title":"G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences.","volume":"39","author":"F Faul","year":"2007","journal-title":"Behav Res Methods."},{"key":"ref22","article-title":"pwr: Basic Functions for Power Analysis.","author":"S Champely","year":"2020","journal-title":"R package version 1.3-0."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"279-282","DOI":"10.4300\/JGME-D-12-00156.1","article-title":"Using Effect Size-or Why the P Value Is Not Enough.","volume":"4","author":"G Sullivan","year":"2012","journal-title":"J Grad Med Educ."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"706-711","DOI":"10.1097\/ACM.0b013e318217e119","article-title":"Does simulation-based medical education with deliberate practice yield better results than traditional clinical education? A meta-analytic comparative review of the evidence.","volume":"86","author":"W McGaghie","year":"2011","journal-title":"Acad Med."},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.4324\/9780203887332","author":"J Hattie","year":"2009","journal-title":"Visible Learning: A Synthesis of Over 800 Meta-Analyses Relating to Achievement."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"582-592","DOI":"10.1097\/01.MLR.0000062554.74615.4C","article-title":"Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation.","volume":"41","author":"G Norman","year":"2003","journal-title":"Med Care."},{"key":"ref27","doi-asserted-by":"publisher","first-page":"599-603","DOI":"10.1097\/EDE.0000000000000876","article-title":"Planning Study Size Based on Precision Rather Than Power.","volume":"29","author":"K Rothman","year":"2018","journal-title":"Epidemiology."},{"key":"ref28","doi-asserted-by":"publisher","first-page":"307-312","DOI":"10.1111\/j..2002.384.doc.x","article-title":"Design and analysis of pilot studies: recommendations for good practice.","volume":"10","author":"G Lancaster","year":"2004","journal-title":"J Eval Clin Pract."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"484-489","DOI":"10.1001\/archpsyc.63.5.484","article-title":"Caution regarding the use of pilot studies to guide power calculations for study proposals.","volume":"63","author":"H Kraemer","year":"2006","journal-title":"Arch Gen Psychiatry."},{"key":"ref30","doi-asserted-by":"publisher","first-page":"1051-1067","DOI":"10.1093\/ije\/dyv113","article-title":"Methods for sample size determination in cluster randomized trials.","volume":"44","author":"C Rutterford","year":"2015","journal-title":"Int J Epidemiol."},{"key":"ref31","doi-asserted-by":"publisher","first-page":"607-610","DOI":"10.1177\/001316447003000308","article-title":"Determining sample size for research activities.","volume":"30","author":"R Krejcie","year":"1970","journal-title":"Educ Psychol Meas."},{"key":"ref32","doi-asserted-by":"publisher","first-page":"217-228","DOI":"10.3109\/0142159X.2015.1105945","article-title":"Improving response rates and evaluating nonresponse bias in surveys: AMEE Guide No. 102.","volume":"38","author":"A Phillips","year":"2016","journal-title":"Med Teach."},{"key":"ref33","doi-asserted-by":"publisher","first-page":"e20476","DOI":"10.1371\/journal.pone.0020476","article-title":"Using and reporting the Delphi method for selecting healthcare quality indicators: a systematic review.","volume":"6","author":"R Boulkedid","year":"2011","journal-title":"PLoS One."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"401-409","DOI":"10.1016\/j.jclinepi.2013.12.002","article-title":"Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies.","volume":"67","author":"I Diamond","year":"2014","journal-title":"J Clin Epidemiol."},{"key":"ref35","doi-asserted-by":"publisher","first-page":"e3","DOI":"10.5395\/rde.2024.49.e3","article-title":"Sample size determination for conducting a pilot study to assess reliability of a questionnaire.","volume":"49","author":"M Bujang","year":"2024","journal-title":"Restor Dent Endod."},{"key":"ref36","doi-asserted-by":"publisher","first-page":"59-82","DOI":"10.1177\/1525822X05279903","article-title":"How many interviews are enough? An experiment with data saturation and variability.","volume":"18","author":"G Guest","year":"2006","journal-title":"Field Methods."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"114523","DOI":"10.1016\/j.socscimed.2021.114523","article-title":"Sample sizes for saturation in qualitative research: A systematic review of empirical tests.","volume":"292","author":"M Hennink","year":"2022","journal-title":"Soc Sci Med."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"84-99","DOI":"10.1037\/1082-989X.4.1.84","article-title":"Sample size in factor analysis.","volume":"4","author":"R MacCallum","year":"1999","journal-title":"Psychol Methods."},{"key":"ref39","doi-asserted-by":"publisher","first-page":"539-549","DOI":"10.1007\/s11136-010-9606-8","article-title":"The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study.","volume":"19","author":"L Mokkink","year":"2010","journal-title":"Qual Life Res."},{"key":"ref40","doi-asserted-by":"publisher","first-page":"1147-1157","DOI":"10.1007\/s11136-018-1798-3","article-title":"COSMIN guideline for systematic reviews of patient-reported outcome measures.","volume":"27","author":"C Prinsen","year":"2018","journal-title":"Qual Life Res."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"2594-2604","DOI":"10.1001\/jama.2012.87802","article-title":"Reporting of noninferiority and equivalence randomized trials: extension of the CONSORT 2010 statement.","volume":"308","author":"G Piaggio","year":"2012","journal-title":"JAMA."},{"key":"ref42","doi-asserted-by":"publisher","first-page":"355-362","DOI":"10.1177\/1948550617697177","article-title":"Equivalence Tests: A Practical Primer for t Tests, Correlations, and Meta-Analyses.","volume":"8","author":"D Lakens","year":"2017","journal-title":"Soc Psychol Personal Sci."},{"key":"ref43","doi-asserted-by":"publisher","first-page":"493-498","DOI":"10.1111\/2041-210X.12504","article-title":"SIMR: an R package for power analysis of generalized linear mixed models by simulation.","volume":"7","author":"P Green","year":"2016","journal-title":"Methods Ecol Evol."},{"key":"ref44","doi-asserted-by":"publisher","first-page":"94","DOI":"10.1186\/1471-2288-11-94","article-title":"Simulation methods to estimate design power: an overview for applied research.","volume":"11","author":"B Arnold","year":"2011","journal-title":"BMC Med Res Methodol."},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1093\/oso\/9780198524847.001.0001","author":"P Diggle","year":"2002","journal-title":"Analysis of Longitudinal Data."},{"key":"ref46","doi-asserted-by":"publisher","first-page":"29","DOI":"10.1186\/s12916-018-1017-7","article-title":"Adaptive designs in clinical trials: why use them, and how to run and report them.","volume":"16","author":"P Pallmann","year":"2018","journal-title":"BMC Med."},{"key":"ref47","doi-asserted-by":"publisher","first-page":"128-142","DOI":"10.3758\/s13423-017-1230-y","article-title":"Bayes factor design analysis: Planning for compelling evidence.","volume":"25","author":"F Sch\u00f6nbrodt","year":"2018","journal-title":"Psychon Bull Rev."},{"key":"ref48","doi-asserted-by":"publisher","first-page":"2134-2156","DOI":"10.1111\/1475-6773.12117","article-title":"Achieving integration in mixed methods designs-principles and practices.","volume":"48","author":"M Fetters","year":"2013","journal-title":"Health Serv Res."},{"key":"ref49","doi-asserted-by":"publisher","first-page":"b2393","DOI":"10.1136\/bmj.b2393","article-title":"Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls.","volume":"338","author":"J Sterne","year":"2009","journal-title":"BMJ."}],"container-title":["F1000Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/f1000research.com\/articles\/15-1147\/v2\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/f1000research.com\/articles\/15-1147\/v2\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/f1000research.com\/articles\/15-1147\/v2\/iparadigms","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T05:34:08Z","timestamp":1789018448000},"score":1,"resource":{"primary":{"URL":"https:\/\/f1000research.com\/articles\/15-1147\/v2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,3]]},"references-count":49,"URL":"https:\/\/doi.org\/10.12688\/f1000research.185853.2","relation":{"has-version":[{"id-type":"doi","id":"10.12688\/f1000research.185853.1","asserted-by":"subject"}],"has-review":[{"id-type":"doi","id":"10.5256\/f1000research.208637.r517508","asserted-by":"subject"},{"id-type":"doi","id":"10.5256\/f1000research.208637.r517508","asserted-by":"object"}]},"ISSN":["2046-1402"],"issn-type":[{"value":"2046-1402","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,3]]},"assertion":[{"value":"Indexed","URL":"https:\/\/f1000research.com\/articles\/15-1147\/v2#article-reports","order":0,"name":"referee-status","label":"Referee status","group":{"name":"current-referee-status","label":"Current Referee Status"}},{"value":"10.5256\/f1000research.205135.r504180, Sushma Prabhath, Manipal Academy of Higher Education, Manipal, Karnataka, 28 Jul 2026, version 1, 3 approved with reservations","URL":"https:\/\/f1000research.com\/articles\/15-1147\/v1#referee-response-504180","order":0,"name":"referee-response-504180","label":"Referee Report","group":{"name":"article-reports","label":"Article Reports"}},{"value":"<b>Anusha Divvi<\/b>; \n<i>Posted: 14 Aug 2026<\/i>; We thank the reviewer for a detailed and constructive report. \n<b>1. Clarify the methodology used to develop this guide.<\/b> A new section, \n<i>Methods: how this guide was developed<\/i>, has been added between Objectives and Main body. It states the three literature streams drawn on; the databases (MEDLINE\/PubMed and Google Scholar), the search terms, and the use of forward and backward citation chaining; three explicit inclusion criteria; and an acknowledgement that no formal quality appraisal and no dual independent selection were performed. It closes by stating that the guide &quot;has not been subjected to a Delphi or other formal consensus process, and it does not carry the authority of a consensus statement.&quot; \n<b>2. Some recommendations are presented too prescriptively.<\/b> Accepted in full; this was the most consistent point across all three reports. It is addressed at four levels. The abstract Conclusions now state that &quot;the numerical benchmarks reproduced here are illustrative anchors rather than methodological minima.&quot; Table 2 has been retitled \n<i>&quot;Commonly cited sample size anchors for designs and methods in which a power-based calculation does not apply&quot;<\/i>, its column heading changed from &quot;Minimum or typical sample&quot; to &quot;Commonly cited anchor&quot;, and a standing caveat added beneath it. Each benchmark you named has been individually softened: the 12-per-group pilot rule is now &quot;an order-of-magnitude anchor rather than a requirement&quot;; the 30 for Cronbach's \u03b1 is &quot;commonly cited as providing reasonable precision&quot; and should be &quot;derived for the specific expected value rather than adopted as a fixed figure&quot;; the 10:1 EFA rule is &quot;the weakest of the available justifications&quot;; and the 9\u201320 interview range now carries &quot;these ranges describe what has been observed empirically in particular study contexts; they are not targets to be met.&quot; The new Limitations section states that applying these anchors outside their originating context &quot;would repeat the error the guide is intended to correct.&quot; \n<b>3. More discussion of software implementation.<\/b> A new section, \n<i>Software for sample size calculation<\/i>, has been added with new Table 3, mapping thirteen design types to G*Power 3.1 menu paths, R functions, and commercial equivalents (PASS, Stata, SAS, Mplus). All five packages you named are covered, and simr is included for simulation-based power. The section names the three errors we most often see in software-based calculation: selecting an independent-samples menu for paired data, entering a between-group \n<i>d<\/i> where the paired \n<i>d<\/i>z is required, and accepting a default one-sided test unintentionally and notes that software executes Step 7 only. Reporting the software, version and procedure is now item 13 of Table 4 and is written into Step 9. \n<b>4. Limited discussion of complex contemporary designs.<\/b> A new section, \n<i>Designs and approaches beyond the scope of this guide<\/i>, covers all seven classes you listed: non-inferiority and equivalence; multilevel and mixed-effects; longitudinal cohort; adaptive designs; Bayesian approaches; and mixed methods, with the explicit statement that &quot;a single sample size justification covering both strands is not appropriate.&quot; Each carries a signpost to the primary literature, and the section closes by stating that the guide is orientation rather than a substitute for specialist consultation. \n<b>5. Greater emphasis on estimation rather than significance testing.<\/b> A new subsection, \n<i>Planning for precision rather than power<\/i>, has been added, with two worked examples and a new row in Table 1. It gives n = (Z\u2081\u208b\u03b1\/\u2082 &times; \u03c3\/E)&sup2; for a single mean \u2014 97 participants to estimate a mean examination score to within &plusmn;2 marks where \u03c3 = 10 \u2014 and n per group = 2Z\u2081\u208b\u03b1\/\u2082&sup2;\/E_d&sup2; for a standardised mean difference, yielding 193 per group to estimate an SMD to within &plusmn;0.20, &quot;appreciably more than the 63 per group required merely to detect \n<i>d<\/i> = 0.5.&quot; Item 4 of the new checklist asks authors to declare whether the study is sized to detect or to estimate. \n<b>6. Table 1 could be improved.<\/b> Table 1 has been expanded from four columns to six and from nine rows to twelve. \n<i>Primary analysis<\/i> is now column 3 and \n<i>Key assumption and common pitfall<\/i> column 6, with a standing assumptions note beneath the table. Software has been placed in a separate table rather than a fourth new column, because the G*Power path, R function and commercial equivalents cannot sit legibly in a Table 1 cell and Table 1 only works as a quick reference if it stays readable; Table 3 is cross-referenced from Table 1's parent section and from Step 7. Three rows have been added: analytical cross-sectional surveys, Delphi and consensus studies, and precision-based planning &nbsp;so the table now covers every design discussed in the text. \n<b>7. Effect size section.<\/b> A new subsection, \n<i>Educational importance, statistical significance, and the credibility of published effect sizes<\/i>, addresses all three points. On minimally important differences: there is no HPE-wide equivalent of the MID, but the half-standard-deviation finding is offered as a starting anchor, and authors are asked to state the difference they target and its basis. On educational versus statistical significance: &quot;a statistically significant improvement of one percentage point on a summative examination in a large cohort may have no implication for curriculum design, while a non-significant difference of a full grade in a small cohort may be highly consequential and merely underpowered.&quot; On publication bias: &quot;in a low-powered literature, the effects that reach significance are systematically the overestimates,&quot; with the recommendation to prefer a pooled estimate or, failing that, the lower bound of the reported confidence interval. \n<b>8. Survey studies.<\/b> Restructured around the three purposes you distinguish, each with its own subheading. \n<i>Estimation surveys<\/i> retain the proportion formula and finite population correction. \n<i>Analytical cross-sectional surveys<\/i> are new, powered using the two-group formulas or the regression model actually to be fitted, with the note that a survey powered only for overall prevalence &quot;will usually be underpowered for the comparisons the investigators intend to report.&quot; \n<i>Psychometric surveys<\/i> are directed to the validation section. A corresponding row has been added to Table 1 and a branch to Figure 2. \n<b>9. Instrument validation.<\/b> Substantially rewritten. Requirements &quot;depend far more on the strength of the factor loadings, the communalities of the measured variables, the number of factors extracted, and the ratio of variables to factors than on the number of participants per item.&quot; Where communalities are high and factors well determined, samples well below 10:1 can recover the structure; where communalities are low, 300 or more may be needed regardless of item count. COSMIN coverage has been strengthened, and the closing sentence asks authors to name the measurement property, state the precision or model-recovery requirement, and cite COSMIN &quot;rather than a generic participant-per-item ratio.&quot; \n<b>10. Attrition.<\/b> Retitled \n<i>Adjusting for expected attrition and missing data<\/i> and extended with three subsections, framed by &quot;inflation protects statistical power; it does not protect validity.&quot; \n<i>Differential attrition<\/i> notes that &quot;the resulting bias is not corrected by recruiting more participants.&quot; \n<i>Missing data mechanisms<\/i> distinguishes MCAR, MAR and MNAR. \n<i>Multiple imputation and analysis-stage implications<\/i> specifies what an imputation plan must state. The section has also been moved to precede the reporting section, correcting a forward reference in Version 1. \n<b>11. Reporting checklist.<\/b> New Table 4, fifteen items in three groups: 1\u201310 for all designs, 11\u201313 covering inflation, missing data and software, and 14\u201315 covering non-power designs and the achieved sample. A prose reporting template is supplied in \n<i>Extended data<\/i>, File S1. \n<b>12. Figures.<\/b> Two figures created; Version 1 had none. Figure 1 presents the nine-step workflow grouped into four labelled phases. Figure 2 is a decision tree routing the reader from study aim through the unit of analysis to the applicable method, with a common tail for attrition inflation and reporting and a flagged branch for designs needing specialist input. We combined the decision tree and design-to-formula algorithm deliberately, since presented separately they would have duplicated the same branching structure. \n<b>Additional suggestions.<\/b> Effect-size discussion has been consolidated into Foundational principles, with Step 4 shortened to a cross-reference and the repetition in the two-group and SBE sections trimmed. Long paragraphs have been divided throughout. A new closing paragraph in Conclusions addresses both future directions you named: simulation-based design analysis, and the use of large language models in study design, which &quot;can generate plausible-looking calculations without the design reasoning that makes a calculation defensible.&quot;","URL":"https:\/\/f1000research.com\/articles\/15-1147\/v2#referee-comment-16930","order":1,"name":"referee-comment-16930","label":"Referee Comment","group":{"name":"article-reports","label":"Article Reports"}},{"value":"10.5256\/f1000research.208637.r517508, Vivek Harsukhbhai Ramanandi, Faculty of Physiotherapy, Marwadi University, Rajkot, Rajkot, 10 Sep 2026, version 2, indexed","URL":"https:\/\/f1000research.com\/articles\/15-1147\/v2#referee-response-517508","order":2,"name":"referee-response-517508","label":"Referee Report","group":{"name":"article-reports","label":"Article Reports"}},{"value":"10.5256\/f1000research.205135.r504181, Manjush Karthika, Liwa University, Abu Dhabi, 07 Aug 2026, version 1, 3 approved with reservations","URL":"https:\/\/f1000research.com\/articles\/15-1147\/v1#referee-response-504181","order":3,"name":"referee-response-504181","label":"Referee Report","group":{"name":"article-reports","label":"Article Reports"}},{"value":"<b>Anusha Divvi<\/b>; \n<i>Posted: 14 Aug 2026<\/i>; We thank the reviewer for a report focused on practical usability, which prompted two of the changes we think most improve the guide. \n<b>1. Inconsistency between the abstract and the effect size section regarding pilot studies.<\/b> Confirmed, and a real internal contradiction. Resolved in the direction you suggest, in three places. The abstract now reads: &quot;we emphasise that the expected difference should be anchored to previous evidence and to educational importance, with pilot work used to estimate variability, recruitment and attrition rather than the effect size itself.&quot; Step 4 no longer presents pilots and published literature as parallel options, stating that pilots &quot;are too small to provide stable effect-size point estimates; their point estimates should not be used directly to power a definitive trial.&quot; The pilot section now gives the reason Version 1 asserted without explaining: &quot;because the observed pilot effect is as likely to be an overestimate as an underestimate and the resulting sample size is inversely proportional to its square, using pilot point estimates systematically underestimates the required sample.&quot; Figure 1, Step 4 also carries &quot;Not a pilot point estimate.&quot; \n<b>2. Table 1 and the worked examples \u2014 state the assumptions and name the software.<\/b> A standing assumptions note now appears beneath Table 1: &quot;\u03b1 = 0.05, two-sided; power = 0.80; 1:1 allocation; complete data with no attrition inflation applied; no adjustment for multiplicity; simple random sampling for survey rows; and, for the two-group means rows, equal variances in the two groups.&quot; A second note names the software and reports the small differences between closed-form and software results rather than leaving a reader who runs G*Power to discover them unexplained; the full reconciliation is given in our response to Reviewer 2, technical point 3. Table 1 also carries a \n<i>Key assumption and common pitfall<\/i> column. Full settings and syntax are in \n<i>Extended data<\/i>, File S1 (https:\/\/doi.org\/10.5281\/zenodo.21929103). \n<b>3a. Survey section \u2014 completed responses versus invitations.<\/b> A real gap, and correcting it required softening an overstatement. Version 1 said flatly that &quot;response rate is a practical consideration in recruitment, not a parameter in sample size calculation,&quot; which is true of the precision calculation but not of the recruitment plan, and as written it discouraged the calculation you are asking for. Version 2 states that &quot;the calculated sample size is the number of \n<i>completed, analysable responses<\/i> required, not the number of people to invite,&quot; with the worked figure: &quot;to obtain 384 completed responses at an anticipated response rate of 40%, approximately 960 invitations are required.&quot; Step 8 is retitled &quot;Inflate for expected attrition and, in surveys, for non-response,&quot; and item 11 of the checklist asks for the number of invitations. \n<b>3b. Limitations section.<\/b> A new \n<i>Limitations<\/i> section has been added before Conclusions, covering: that this is a practical guide rather than a systematic review or consensus statement; that the numerical anchors are illustrative; that the closed-form formulas apply to simple, balanced, single-outcome cases only; and, as you ask, that complex designs require specialised support &quot;no guide substitutes for methodological expertise. Investigators planning studies with complex designs, unusual outcome structures, or high-stakes implications should involve a statistician or methodologist at the design stage, not after data collection.&quot; \n<b>3c. Language, abbreviations, equation formatting, and &quot;New York&quot;.<\/b> The affiliation is corrected. Notation is standardised. Every abbreviation is defined at first use, and HPE, SBE, COSMIN, SAMPL, ICC, DE, EFA, CFA, SEM and SMD are also defined in the figure footnotes so the figures stand alone. All display equations have been reset on their own lines, including the unequal-allocation formula, which contained a genuine error. \n<b>4. Reporting checklist.<\/b> New Table 4 provides fifteen items. Every element you specified is included: primary outcome (1), effect size source (6), \u03b1 and power (7), allocation ratio (8), attrition adjustment and recruitment target (11), and software (13) with six further items the three reports collectively implied: unit of analysis (2), analytical model (3), detect-versus-estimate (4), test direction (7), missing data approach (12), and achieved sample (15). A prose reporting template is given at the end of File S1. \n<b>5. Software examples with exact settings or syntax in supplementary material.<\/b> Addressed in full; File S1 contains nine worked examples with G*Power field lists, runnable R syntax and expected output, including the simr route for cluster designs, which G*Power does not support. It also flags the specific input errors that produce wrong answers in G*Power, notably entering a between-group \n<i>d<\/i> where \n<i>d<\/i>z is required. \n<b>6. Visual workflow.<\/b> Two figures created; Version 1 had none. Figure 1 presents the nine-step workflow grouped into four labelled phases. Figure 2 is a decision tree routing the reader from study aim through the unit of analysis to the applicable method, with a common tail for attrition inflation and reporting and a flagged branch for designs needing specialist input. We combined the decision tree and design-to-formula algorithm deliberately, since presented separately they would have duplicated the same branching structure.","URL":"https:\/\/f1000research.com\/articles\/15-1147\/v2#referee-comment-16932","order":4,"name":"referee-comment-16932","label":"Referee Comment","group":{"name":"article-reports","label":"Article Reports"}},{"value":"The author(s) declared that no grants were involved in supporting this work","order":5,"name":"grant-information","label":"Grant Information"},{"value":"This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.","order":0,"name":"copyright-info","label":"Copyright"}]}}