On the use of a pilot sample for sample size determination
The DOI matches its Crossref record; 9 of the 9 references it cites were found in a registry or an index.
Abstract
To compute the sample size needed to achieve the planned power for a t ‐test, one needs an estimate of the population standard deviation δ. If one uses the sample standard deviation from a small pilot study as an estimate of δ, it is quite likely that the actual power for the planned study will be less than the planned power. Monte Carlo simulations indicate that using a 100(1 − γ) per cent upper one‐sided confidence limit on δ will provide a sample size sufficient to achieve the planned power in at least 100(1 − γ) per cent of such trials.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Consistent with recommended pilot sample sizes derived from commonly-used approaches for pilot and feasibility trials [28–30], the target sample size was N=60 (30 per group).”
In the citing paper
“Most methodological papers that focus on recommendations about sample size requirements for pilot trials assume that the main aim of such a trial is to estimate a quantitative measure such as the variance (or standard deviation) of an effect size to inform the sample size calculation for a future definitive RCT.”
From CONSORT 2010 statement: extension to randomised pilot and feasibility trials. (2016) · cites this paper as [57]
In the citing paper
“There is no consensus but recommendations vary from 10 to 12 per group through to 60–75 per group depending on the main objective of the study.”
From Determining sample size for progression criteria for pragmatic pilot RCTs: the hypothesis test strikes back! (2021) · cites this paper as [8]
In the citing paper
“a minimum of 30 participants was considered enough to reach the aims of feasibility and to provide useful data to optimize a future RCT design.”
From Feasibility, acceptability, and preliminary efficacy of a blended transdiagnostic group CBT for the treatment of emotional disorders (2025) · cites this paper as Browne, 1995
In the citing paper
“there is growing evidence for the effectiveness of mindfulness courses for people with long-term medical conditions.”
From Distant delivery of a mindfulness-based intervention for people with Parkinson's disease: The study protocol of a randomised pilot trial (2017) · cites this paper as [23]
4 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Statistical principles in experimental design.
Winer, B. J. (1962) · McGraw-Hill Book Company · doi:10.1037/11774-000 · Cited by 26,755
CONSORT statement: extension to cluster randomised trials
Campbell, M. K., Elbourne, D. R., & Altman, D. G. (2004) · BMJ · doi:10.1136/bmj.328.7441.702 · Cited by 2,557
Statistical Principles in Experimental Design.
Geisser, S., & Winer, B. J. (1963) · Journal of the American Statistical Association · doi:10.2307/2283359 · Cited by 1,890
Statistical Power Analysis for the Behavioral Sciences
Wassertheil, S., & Cohen, J. (1970) · Biometrics · doi:10.2307/2529115 · Cited by 1,383
Statistical Methods in Epidemiology
(2012) · SpringerReference · doi:10.1007/springerreference_205629 · Cited by 1,124
STATISTICAL METHODS IN EPIDEMIOLOGY
Yu, S., & Wang, X. (2017) · Handbook of Medical Statistics · doi:10.1142/9789813148963_0018 · Cited by 224
A Correction for Biasing Effects of Pilot Sample Size on Sample Size Determination
Shiffler, R. E., & Adams, A. J. (1987) · Journal of Marketing Research · doi:10.1177/002224378702400309 · Cited by 8
An Empirical Assessment of Realized A-Risk when Testing Hypotheses'
Shiffler, R. E., & Harwood, G. B. (1985) · Educational and Psychological Measurement · doi:10.1177/0013164485454012 · Cited by 3
A Correction for Biasing Effects of Pilot Sample Size on Sample Size Determination
Shiffler, R. E., & Adams, A. J. (1987) · Journal of Marketing Research · doi:10.2307/3151643 · Cited by 2
Research works printed but not linked: 1
- Cochran W. G., 1977, Sampling Techniques
Cited by
Items 201–300 of 997
Shin, D. W., Yoon, H., Kim, H. S., Choi, Y. J., Shin, C. M., Park, Y. S., Kim, N., & Lee, D. H. (2018) · Evidence-based Complementary and Alternative Medicine · doi:10.1155/2018/9147914 · Cited by 29
“This study was planned with 30 participants per group, which is the minimum number of participants recommended in a pilot study [25, 26].”
BackgroundKyte, D. G., Bishop, J. R. B., Brettell, E. A., Calvert, M., Cockwell, P., Dutton, M., Eddington, H., Hadley, G., Ives, N. J., Jackson, L. J., et al. (2018) · BMJ Open · doi:10.1136/bmjopen-2018-026080 · Cited by 24
Shin, H.-Y., Kim, H.-R., Jahng, G., Jin, C., Kwon, S., Cho, S., Park, S.-U., Jung, W., Moon, S., Ko, C., et al. (2021) · BMC Complementary Medicine and Therapies · doi:10.1186/s12906-021-03428-6 · Cited by 28
“Furthermore, Browne suggested a general flat rule to use at least 30 subjects to estimate a parameter [36].”
BackgroundDean, S. G., Poltawski, L., Förster, A., Taylor, R. S., Spencer, A. E., James, M. A., Allison, R., Stevens, S., Norris, M., Shepherd, A. I., et al. (2016) · BMJ Open · doi:10.1136/bmjopen-2016-012375 · Cited by 25
“This number is based on the recommendation of 30 complete data sets for pilot studies in order to estimate outcome variance 27 and running the intervention three times (Exeter I and South Devon, then Exeter II, ie, 3×8 patients) to enable investigation of variations in context and other process…”
MethodFacet-joint injections for non-specific low back pain: a feasibility RCT
Snidvongs, S., Taylor, R. S., Ahmad, A., Thomson, S., Sharma, M. L., Farr, A., Fitzsimmons, D. F., Poulton, S., Mehta, V., & Langford, R. (2017) · Health Technology Assessment · doi:10.3310/hta21740 · Cited by 23
Hulin, J., Baker, S. R., Marshman, Z., Albadri, S., & RODD, H. D. (2016) · International Journal of Paediatric Dentistry · doi:10.1111/ipd.12267 · Cited by 32
Campo, Y. E. D., Cubillos, L., Vu, M. B., Aguirre, A. R., Reuland, D. S., & Keyserling, T. C. (2017) · Ethnicity and Health · doi:10.1080/13557858.2017.1346784 · Cited by 27
“For the Implementation phase, a sample size of 40 was selected to assess the feasibility and acceptability of the intervention, and to get estimates of variance for outcome measures, anticipating up to a 20% attrition rate (Browne 1995; Lancaster, Dodd, and Williamson 2004; Sim and Lewis 2012).”
Determinants of choice behaviour of Islamic investment products in Malaysia
Amin, H., Suhartanto, D., Ali, M., Ghazali, M. F., Rizal, H., & Razak, D. A. (2022) · Journal of Islamic marketing · doi:10.1108/jima-09-2021-0302 · Cited by 28
Morita Therapy for depression (Morita Trial): a pilot randomised controlled trial
Sugg, H. V., Richards, D., & Frost, J. (2018) · BMJ Open · doi:10.1136/bmjopen-2018-021605 · Cited by 23
Findings from a pilot randomised trial of a social network self-management intervention in COPD
Welch, L., Orlando, R., Lin, S. X., Vassilev, I., & Rogers, A. E. (2020) · BMC Pulmonary Medicine · doi:10.1186/s12890-020-1130-1 · Cited by 24
“CAT 21 [16, 26] 21 [16, 24] -1 [−5, 2] 18 [10, 22] 15 [10, 25] 0 [−3, 1] 349 p = 0.”
BackgroundRobinson, L. E., Wang, L., Colabianchi, N., Stodden, D. F., & Ulrich, D. A. (2020) · BMJ Open · doi:10.1136/bmjopen-2020-037497 · Cited by 30
Using the Sample Range as a Basis for Calculating Sample Size in Power Calculations
Browne, R. H. (2001) · The American Statistician · doi:10.1198/000313001753272420 · Cited by 22
“Recommendations on how best to use a small-sample estimate of a to compute N were given by Browne (1995). However, a sample standard deviation is not always readily available.”
MethodSelf-citationHadlandsmyth, K., Conrad, M., Steffensmeier, K. S., Van Tiem, J., Obrecht, A., Cullen, J. J., & Vander Weg, M. W. (2022) · Annals of Surgery · doi:10.1097/sla.0000000000004671 · Cited by 21
“Initially, targeted enrollment was N = 75 with the goal of 30 per arm (20% attrition) to provide pilot data to estimate effect sizes for a subsequent larger trial [27-29].”
MethodSingh, S., Avinash, R., Jaiswal, S., & Kumari, A. (2023) · Indian Journal of Anaesthesia · doi:10.4103/ija.ija_844_21 · Cited by 21
“Additionally, Browne[10] cites a general flat rule to ‘use at least 30 subjects or greater to estimate a parameter’.”
BackgroundExtracorporeal membrane oxygenation (ECMO) reconsidered
Moran, J. L., Chalwin, R. P., & Graham, P. L. (2010) · Critical Care and Resuscitation · doi:10.1016/s1441-2772(23)01516-8 · Cited by 25
Indira, V., Vasanthakumari, R., & Sugumaran, V. (2010) · Expert Systems with Applications · doi:10.1016/j.eswa.2010.06.068 · Cited by 23
Evaluation of accuracy of capillary refill index with pneumatic fingertip compression
Shinozaki, K., Saeki, K., Jacobson, L. S., Falotico, J. M., Li, T., Hirahara, H., Horie, K., Kobayashi, N., Weisner, S., Lampe, J. W., et al. (2020) · Journal of Clinical Monitoring and Computing · doi:10.1007/s10877-019-00454-1 · Cited by 24
Deschamps, T., Sauvaget, A., Pichot, A., Valrivière, P., Maroulidès, M., Bois, A., Bulteau, S., & Thomas-Ollivier, V. (2016) · Journal of Psychiatric Research · doi:10.1016/j.jpsychires.2016.08.013 · Cited by 25
“current perspective, these positive results lead to a trial protocol for administering rTMS treatment among depressed patients with appropriate sample size (Browne, 1995), and for testing”
BackgroundO'Connor, K. A., Bagnell, A. L., McGrath, P. J., Wozney, L., Radomski, A. D., Rosychuk, R. J., Curtis, S. J., Jabbour, M., Fitzpatrick, E. A., Johnson, D. W., et al. (2020) · JMIR Mental Health · doi:10.2196/13356 · Cited by 22
“A two-sided two-sample t test power analysis was conducted for the change in score from baseline to postintervention to calculate the sample required per group in a definitive RCT [40].”
MethodDoes improving indoor air quality lessen symptoms associated with chemical intolerance?
Perales, R. B., Palmer, R. F., Rincon, R., Viramontes, J. N., Walker, T., Jaén, C. R., & Miller, C. S. (2022) · Primary Health Care Research & Development · doi:10.1017/s1463423621000864 · Cited by 23
“Sample sizes of 20–30 with a minimum of 12 for pilot studies have been recommended by various research statisticians (Birkett & Day, 1994; Browne, 1995).”
BackgroundPsychological mobile app for patients with acute myeloid leukemia: A pilot randomized clinical trial
El-Jawahri, A. R., Luskin, M. R., Greer, J. A., Traeger, L. N., Lavoie, M. W., Vaughn, D., Andrews, S., Yang, D. X., Boateng, K. Y., Newcomb, R. A., et al. (2023) · Cancer · doi:10.1002/cncr.34645 · Cited by 26
Aminuddin, F., Bahari, M. S., Zainuddin, N. A., Hanafiah, A. N. M., & Hassan, N. Z. A. M. (2023) · Asian Pacific Journal of Cancer Prevention · doi:10.31557/apjcp.2023.24.2.489 · Cited by 26
“The sample size was selected in accordance with a general flat rule for pilot feasibility studies (Lancaster et al., 2004; Browne, 1995).”
MethodKeihanian, T., Zabad, N., Khalaf, M. A., Abdel-Jaber, W., Kim, Y., Jawaid, S. A., Abidi, W. M., Marginean, E. C., & Othman, M. O. (2023) · Gastrointestinal Endoscopy · doi:10.1016/j.gie.2023.04.006 · Cited by 20
Bias in linear model power and sample size due to estimating variance
Muller, K. E., & Pasour, V. (1997) · Communication in Statistics- Theory and Methods · doi:10.1080/03610929708831953 · Cited by 20
“Browne (1995) and many others (cited in Taylor and Muller, 1995) have reported approximate solutions.”
BackgroundDwyer, L., Parkin-Smith, G. F., Brantingham, J. W., Korporaal, C., Cassa, T. K., Globe, G. A., Bonnefin, D., & Tong, V. (2014) · Journal of Manipulative and Physiological Therapeutics · doi:10.1016/j.jmpt.2014.10.002 · Cited by 25
“”(44) The authors agreed on a minimum sample size of ≈25 per group (3 groups), while considering available funding, practical constraints, and the available time frame for recruitment.(44,45)”
BackgroundWilkinson, E. L., O’Mahen, H., Fearon, P. R. M., Halligan, S. L., King, D. X., Greenfield, G., Dunkley-Bent, J., Ericksen, J., Milgrom, J., & Ramchandani, P. G. (2016) · Trials · doi:10.1186/s13063-016-1274-8 · Cited by 22
“A sample size of 30 participants per arm is recommended for feasibility studies as providing sufficient data to gain an accurate estimation of the feasibility and acceptability of the intervention and trial methods [40], and to provide a reasonable range of estimates of the sample size required for a definitive RCT [4…”
BackgroundStrong, S. A., Pető, T., Bunce, C., Xing, W., Georgiou, M., Esposti, S. D., Kalitzeos, A. A., Webster, A. R., & Michaelides, M. (2020) · British Journal of Ophthalmology · doi:10.1136/bjophthalmol-2019-315152 · Cited by 20
Bhargava, M. V., Rana, S. S., Gorsi, U., Kang, M., & Gupta, R. (2021) · Digestive Diseases and Sciences · doi:10.1007/s10620-021-07201-0 · Cited by 20
Increasing Fruit and Vegetable Consumption Through a Healthy Eating Blog: A Feasibility Study
Caplette, M.-E., Provencher, V., Bissonnette-Maheux, V., Dugrenier, M., Lapointe, A., Gagnon, M., Straus, S. E., & Desroches, S. (2017) · JMIR Research Protocols · doi:10.2196/resprot.6622 · Cited by 24
“The sample size for feasibility trials is typically determined pragmatically, with recommendations of a minimum of 30 participants per group [30,31], which is what we aimed for in our study (n=40 per group to account for dropouts).”
Buro, A. W., Gray, H. L., Kirby, R. S., Marshall, J. T., Strange, M., Hasan, S., & Holloway, J. M. (2022) · Journal of Nutrition Education and Behavior · doi:10.1016/j.jneb.2022.01.008 · Cited by 24
Henderson, V., Madrigal, J. M., Kendall, L., Parekh, P., Newsome, J., Chukwudozie, I. B., Comer-HaGans, D., Coffey, V., Grumbach, G., Spencer, S., et al. (2022) · BMC Health Services Research · doi:10.1186/s12913-022-08193-x · Cited by 22
“A sample size of n = 30 is commonly recommended for pilot studies [37].”
BackgroundA pilot with computer-assisted psychosocial risk –assessment for refugees
Ahmad, F., Shakya, Y. B., Li, J., Khoaja, K., Norman, C. D., Lou, W. Y. W., Abuelaish, I., & Ahmadzi, H. M. (2012) · BMC Medical Informatics and Decision Making · doi:10.1186/1472-6947-12-71 · Cited by 19
“As this study was a pilot trial, we aimed to recruit 50 refugees based on “ Recommendations for Good Practice ” by Lancaster et al for pilot randomized trials [22], consistent with others [23].”
MethodLi, Y., Solomon, P., Zhang, A., Franklin, C., Ji, Q., & Chen, Y. (2017) · Health & Social Work · doi:10.1093/hsw/hlx045 · Cited by 24
Oral Microbiota in Severe Early Childhood Caries in Thai Children and Their Families: A Pilot Study
Ledder, R. G., Kampoo, K., Teanpaisan, R., & McBain, A. J. (2018) · Frontiers in Microbiology · doi:10.3389/fmicb.2018.02420 · Cited by 22
“The number of participants (n = 32) is in agreement with previous recommendations (Browne, 1995) for recruiting a representative sample size for a pilot trial.”
ResultKulick, D., Langer, R. D., Ashley, J. M., Gans, K. M., Schlauch, K. A., & Feller, C. (2013) · BMC Family Practice · doi:10.1186/1471-2296-14-59 · Cited by 24
“Statistical methods The sample size was chosen on the theoretical basis that at least 30 subjects per treatment group is reasonable for use of a parametric two-sample t-test to estimate treatment differences, assuming a normal distribution of the outcome measures [13].”
MethodMiller, C. C., Burnside, G., Higham, S. M., & Flannigan, N. (2016) · The Angle Orthodontist · doi:10.2319/092415-648.1 · Cited by 22
“A sample size of 30 was chosen, as this would allow estimation of parameters for a sample size calculation to be conducted in future definitive studies.(11) Consecutive patients attending Liverpool University Dental Hospital for orthodontic treatment conducted by the same clinician were asked to participate.”
MethodYang, S.-Y., Moss-Morris, R. E., & McCracken, L. M. (2017) · Pain Research and Treatment · doi:10.1155/2017/6916915 · Cited by 22
“A sample size of minimum 𝑁 = 30 has been recognized as a reasonable minimum sample size needed for parameter estimates of a larger RCT [58, 59].”
BackgroundHind, D., Parkin, J., Whitworth, V., Rex, S., Young, T. A., Hampson, L. V., Sheehan, J., Maguire, C., Cantrill, H., Scott, E., et al. (2017) · Pilot and Feasibility Studies · doi:10.1186/s40814-017-0132-0 · Cited by 30
“The intended sample of 40 would have permitted power calculations for a future full-scale trial [20, 28].”
BackgroundMethodMitchell, F., Wilkie, L., Robertson, K. J., Reilly, J. J., & Kirk, A. F. (2017) · Pediatric Diabetes · doi:10.1111/pedi.12615 · Cited by 18
“The sample size was based on recommendations from authors specializing in feasibility and pilot studies.(22,23) The sample will introduce sufficient variance to examine the feasibility of a larger study.”
MethodLester, E. G., Macklin, E. A., Plotkin, S. R., & Vranceanu, A. (2020) · Journal of Neuro-Oncology · doi:10.1007/s11060-020-03441-8 · Cited by 18
Maddali, M. M., Al-Shamsi, F., Arora, N. R., & Panchatcharam, S. M. (2020) · Journal of Cardiothoracic and Vascular Anesthesia · doi:10.1053/j.jvca.2020.03.007 · Cited by 18
Wilson, D. T., Walwyn, R. E. A., Brown, J. M., Farrin, A. J., & Brown, S. (2015) · Statistical Methods in Medical Research · doi:10.1177/0962280215589507 · Cited by 18
Jang, S., Lee, J. A., Jang, B.-H., Shin, Y. C., Ko, S., & Park, S. (2019) · The Journal of Alternative and Complementary Medicine · doi:10.1089/acm.2019.0009 · Cited by 21
Ward, J. A., Balian, C., Gilger, E., Raybin, J. L., Li, Z., & Montgomery, K. E. (2020) · Journal of Pediatric Oncology Nursing · doi:10.1177/1043454220917686 · Cited by 18
“The target sample size was estimated at 60 patients and determined based on the projected time period of the study, anticipated attrition, as well as the recommendation of including between 24 and 50 subjects for a pilot study (Browne, 1995).”
Palmieri, F., Gomis, P., Ferreira, D., Ruiz, J. E., Liberal, B. B., Martín-Yebra, A., Bukhari, S. H. A., Pueyo, E., Martínez, J. P., Ramírez, J., et al. (2021) · Scientific Reports · doi:10.1038/s41598-021-82935-5 · Cited by 20
Student and Parent Perspectives on Severe Food Allergies at College
Ersig, A. L., & Williams, J. K. (2018) · Journal of Pediatric Health Care · doi:10.1016/j.pedhc.2018.03.006 · Cited by 21
Soliman, M., Mowafy, K., Elsaadany, N., Soliman, R. M., & Elmetwally, A. M. (2020) · SAGE Open Medicine · doi:10.1177/2050312120927636 · Cited by 17
Parker, S. J., Zhang, A. L., Zhang, C. S., Goodman, G. J., Wen, Z., Yan, Y., Yao, D., Wu, H., Deng, H., Lu, C., et al. (2021) · Clinical and Translational Medicine · doi:10.1002/ctm2.286 · Cited by 18
Game, F. L., Gray, K., Davis, D., Sherman, R., Chokkalingam, K., Connan, Z., Fakis, A., & Jones, M. (2021) · International Wound Journal · doi:10.1111/iwj.13571 · Cited by 23
“A sample size target of 30 participants was chosen as an acceptable sample size for pilot studies.(8,9)”
MethodWilliams, T. I., Loucas, T., Sin, J., Jeremic, M., Aslett, G., Knight, M., Fincham-Majumdar, S., & Liu, F. (2021) · Pilot and Feasibility Studies · doi:10.1186/s40814-021-00918-9 · Cited by 22
“Finally, it has been recommended that pilot studies should include 12 per group [26] or 30 or more participants [27, 28] in order Keywords: Autism spectrum disorder, Language intervention, Music‐assisted programmes (MAP), Treatment‐as‐usual (TAU), Minimally verbal, Telehealth, Parent‐mediated, Social communication”
Kyte, D. G., Anderson, N. E., Bishop, J. R. B., Bissell, A., Brettell, E. A., Calvert, M., Chadburn, M., Cockwell, P., Dutton, M., Eddington, H., et al. (2022) · BMJ Open · doi:10.1136/bmjopen-2021-050610 · Cited by 21
van den Bosch, O. F. C., Gleicher, Y., Arzola, C., Siddiqui, N. T., Downey, K., & Carvalho, J. C. A. (2022) · Regional Anesthesia & Pain Medicine · doi:10.1136/rapm-2022-103948 · Cited by 17
Beyond conventions: Unravelling perceived value's role in shaping digital-only banks' adoption
Saif, M. A. M., Hussin, N., Husin, M. M., Muneer, A., & Alwadain, A. S. (2024) · Technological Forecasting and Social Change · doi:10.1016/j.techfore.2024.123337 · Cited by 43
Toebes, M. J. P., Hoozemans, M. J. M., Mathiassen, S. E., Dekker, J., & van Dieën, J. H. (2015) · Aging Clinical and Experimental Research · doi:10.1007/s40520-015-0390-8 · Cited by 17
“Table 2) rather than on the point estimate, which is in general downward (‘‘optimistically’’) biased [53].”
BackgroundPl, L., & Sy, W. (2012) · The Malaysian journal of psychiatry · Cited by 17
Woodford, J., Farrand, P. A., Watkins, E. R., Richards, D. A., & Llewellyn, D. J. (2014) · Trials · doi:10.1186/1745-6215-15-157 · Cited by 20
“As such, we will use the recommended sample size of 30 participants per arm for feasibility studies [56] consistent with the median sample size found in both feasibility and pilot RCTs [57].”
MethodHurley, D. A., Jeffares, I., Hall, A. M., Keogh, A., Toomey, E. C., McArdle, D., Lonsdale, C., Guérin, S., Guerin, S., Segurado, R., et al. (2020) · Trials · doi:10.1186/s13063-020-04671-x · Cited by 18
“However, a “rule-ofthumb” recommendation for pilot studies is that 30 participants per arm be enrolled to estimate parameters for future sample size calculations [28].”
BackgroundTesting a Brief Self-Directed Behavioral Weight Control Program
Linde, J. A., & Jeffery, R. W. (2011) · Behavioral Medicine · doi:10.1080/08964289.2011.568992 · Cited by 20
“Based on guidelines for determining sample size in a pilot study, 17 the study goal of recruiting a minimum of 30 participants to each condition at baseline was met.”
Bennett, M., Mulvey, M. R., Campling, N. C., Latter, S. M., Richardson, A., Bekker, H., Blenkinsopp, A., Carder, P., Closs, J., Farrin, A. J., et al. (2017) · Health Technology Assessment · doi:10.3310/hta21760 · Cited by 22
Infrared and Raman spectroscopy of blood plasma for rapid endometrial cancer detection
Schiemer, R., Grant, J., Shafiee, M. N., Phang, S., Furniss, D., Boitor, R., Seddon, A. B., Notingher, I., Atiomo, W., Jones, N. W., et al. (2025) · British Journal of Cancer · doi:10.1038/s41416-025-03050-0 · Cited by 18
Mitchell, F., Stevens, G., Jahoda, A. J., Matthews, L., Hankey, C., Murray, H. M., & Melville, C. A. (2018) · Journal of Applied Research in Intellectual Disabilities · doi:10.1111/jar.12489 · Cited by 21
“The sample size was based on recommendations from authors specializing in feasibility and pilot studies (Browne, 1995; Lancaster, Dodd & Williamson, 2004).”
MethodMaurer, T., von Grundherr, J., Patra, S., Jaeger, A., Becher, H., Schmalfeldt, B., Zyriax, B., & Chang-Claude, J. C. (2019) · International Journal of Gynecological Cancer · doi:10.1136/ijgc-2019-000585 · Cited by 16
Schiavo, S., Richardson, D. L., Mina, D. S., Buryk-Iggers, S., Uehling, J. M., Carroll, J., Clarke, H. A., Djaiani, C., Gershinsky, M., & Katznelson, R. (2020) · Applied Physiology Nutrition and Metabolism · doi:10.1139/apnm-2020-0124 · Cited by 16
Effectiveness of Ayurveda treatment in Urdhwaga Amlapitta: A clinical evaluation
Khapre, M. P., Vinteshwari, N., Minaxi, J., & Saxena, V. (2021) · Journal of Ayurveda and Integrative Medicine · doi:10.1016/j.jaim.2020.12.004 · Cited by 16
“Sample size: According to the central limit theorem and Browne's (1995) estimate of sample size for the pilot study, a sample size of 30 was considered [11,12].”
MethodIyota, K., Mizutani, S., Kishimoto, H., Oku, S., Tani, A., Yatsugi, H., Chu, T., Liu, X., & Kashiwazaki, H. (2021) · Gerontology · doi:10.1159/000518270 · Cited by 19
Hamdani, S. U., Huma, Z., Nizami, A. T., Baneen, U.-U., Suleman, N., Javed, H., Malik, A., Wang, D., Mazhar, S., Khan, S. A., et al. (2022) · Child and Adolescent Psychiatry and Mental Health · doi:10.1186/s13034-022-00480-z · Cited by 23
“Since this study was a feasibility cRCT, no power calculations were done; however, the sample was adequate to estimate the parameters for a planned definitive cRCT [5, 28].”
MethodZhang, H., Yoong, S. Q., Dong, Y. H., Goh, S. H., Lim, S., Chan, Y. S., Wang, W., & Wu, X. V. (2022) · Nurse Educator · doi:10.1097/nne.0000000000001294 · Cited by 20
Schneider, J., Matheson, E. L., Tinoco, A., Gentili, C., White, P., Boucher, C., Silva-Breen, H. L., Goorevich, A., Diedrichs, P. C., & LaVoi, N. M. (2023) · Journal of Applied Sport Psychology · doi:10.1080/10413200.2023.2212023 · Cited by 19
“Using Browne ’ s method (Browne, 1995), for an 80% chance of avoiding underpowering a definitive trial and a 70% chance of not excessively overpowering, each arm in the pilot study should have at least n ¼ 16 participants (32 in total) providing outcome data.”
MethodPark, S., Tang, A., Barclay, R., Bayley, M. T., Eng, J. J., MacKay-Lyons, M. J., Pollock, C. L., Pooyania, S., Teasell, R. W., Yao, J. K., et al. (2023) · Physical Therapy · doi:10.1093/ptj/pzad165 · Cited by 18
Sugg, H. V. R., Richards, D. A., & Frost, J. (2016) · Trials · doi:10.1186/s13063-016-1279-3 · Cited by 18
“30 participants per group is also in line with the general rule of thumb for using pilot studies to reliably estimate variance for participant outcomes [26].”
Funes, C. J., Mace, R. A., Macklin, E. A., Plotkin, S. R., Jordan, J. T., & Vranceanu, A. (2019) · Journal of Neuro-Oncology · doi:10.1007/s11060-019-03182-3 · Cited by 16
Hughes, N., Bennett, M., & Johnson, M. I. (2012) · Clinical Journal of Pain · doi:10.1097/ajp.0b013e3182579919 · Cited by 18
Asfar, T., Caban-Martinez, A. J., McClure, L. A., Ruano-Herreria, E. C., Sierra, D., Clark, G. G., Samano, D., Dietz, N. A., Ward, K. D., Arheart, K. L., et al. (2018) · Contemporary Clinical Trials · doi:10.1016/j.cct.2018.02.007 · Cited by 16
Liv, P., Mathiassen, S. E., & Svendsen, S. W. (2012) · BMC Medical Research Methodology · doi:10.1186/1471-2288-12-58 · Cited by 16
“In order to account for variance estimation uncertainty in power analyses, some authors have suggested to use the 80 percentile of the expected distribution of variance estimates as an input rather than the actual variance estimate [36].”
MethodFarrelly, S., Peters, E., Azis, M., David, A., & Hunter, E. C. (2016) · Pilot and Feasibility Studies · doi:10.1186/s40814-016-0086-7 · Cited by 15
“Population variances will be determined using the upper 80th nonparametric bootstrap percentile of confidence intervals around the estimates [42].”
MethodLongden, E., Corstens, D., Pyle, M., Emsley, R., Peters, S., Chauhan, N., Dehmahdi, N., & Morrison, A. P. (2021) · Psychosis · doi:10.1080/17522439.2021.1884740 · Cited by 17
“A target of 50 participants was estimated as sufficient to obtain precise, reliable parameter estimates for sample size in a future definitive trial, as well as demonstrating feasibility (Browne, 1995).”
BackgroundMartínez‐Borba, V., Suso‐Ribera, C., Díaz‐García, A., Salat‐Batlle, J., Castilla, D., Zaragozá, I., García‐Palacios, A., & Sánchez‐Raya, J. (2021) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph18157767 · Cited by 15
“Sample size has been calculated according to previous recommendations suggesting a minimum of 30 participants in feasibility and pilot studies [27] and a minimum of 15 participants in usability studies with devices used in healthcare settings [28].”
MethodBulgarelli, C., Pinti, P., Aburumman, N., & Jones, E. J. H. (2023) · Oxford Open Neuroscience · doi:10.1093/oons/kvad012 · Cited by 24
“Although it is sufficient for pilot and proof-of principle studies [77], future studies might need to recruit more participants to more confidently and reliably assess preschoolers social development, its neural correlates, and any gender-or age-related differences.”
BackgroundPower calculation for overall hypothesis testing with high‐dimensional commensurate outcomes
Chi, Y., Gribbin, M., Johnson, J. L., & Muller, K. E. (2013) · Statistics in Medicine · doi:10.1002/sim.5986 · Cited by 15
“Browne [19], Taylor and Muller [16, 20], and Muller and Pasour [21] described how to account for using estimates in power calculations in various settings with the univariate model.”
BackgroundMcCormick, S. A., McDonald, K. R., Vatter, S., Orgeta, V., Poliakoff, E., Smith, S. J., Silverdale, M., Fu, B., & Leroi, I. (2017) · BMJ Open · doi:10.1136/bmjopen-2017-016801 · Cited by 16
“By enrolling 32 dyads per group, it will allow for a 15% attrition rate (consistent with pilot sample guidance 17 ).”
van Berkel, D., Ong, T. I. W., Drummond, A. E., Hendrick, P. A., Leighton, P. A., Jones, M., Salem, K. M. I., Quraishi, N. A., Brookes, C. L., Paola, A. S. D., et al. (2019) · BMJ Open · doi:10.1136/bmjopen-2019-032111 · Cited by 15
Shinozaki, K., Jacobson, L. S., Saeki, K., Kobayashi, N., Weisner, S., Falotico, J. M., Li, T., Kim, J. H., Lampe, J. W., & Becker, L. B. (2020) · The American Journal of Emergency Medicine · doi:10.1016/j.ajem.2020.04.007 · Cited by 15
Stonbraker, S., Liu, J., Sanabria, G., George, M., Amesty, S. C., Alcántara, C., Abraído‐Lanza, A. F., Halpern, M., Rowell‐Cunsolo, T. L., Henry, S. B., et al. (2021) · AIDS and Behavior · doi:10.1007/s10461-021-03331-8 · Cited by 17
Factors infl uencing consumers’ behaviour towards organic food purchase in Denmark and Tanzania
Pacho, F. T., & Batra, M. M. (2021) · Studies in Agricultural Economics · doi:10.7896/j.2127 · Cited by 20
“We followed a recommendation by Browne (1995) that 30 independent respondents or more are adequate for estimating a parameter.”
MethodAbdullah, K., Thorpe, K. E., Mamak, E. G., Maguire, J. L., Birken, C. S., Fehlings, D. L., Hanley, A. J., Macarthur, C., Zlotkin, S. H., & Parkin, P. C. (2015) · Trials · doi:10.1186/s13063-015-0635-z · Cited by 16
“Application of prior estimates for power calculation of the current trial may lead to an unnecessarily large trial, or the trial may not be large enough to have sufficient power for detection of a clinically relevant treatment effect [66,68].”
BackgroundBowers, B. J., Jacobson, N. A., & Krupp, A. E. (2016) · Research in Nursing & Health · doi:10.1002/nur.21752 · Cited by 14
“There are numerous rules of thumb regarding sample size appropriateness in pilot or exploratory studies (Browne, 1995; Hertzog, 2008; Julious, 2005).”
BackgroundPasanisi, P., Gariboldi, M., Verderio, P., Signoroni, S., Mancini, A., Rivoltini, L., Milione, M., MASCI, E., Ciniselli, C. M., Bruno, E., et al. (2019) · Integrative Cancer Therapies · doi:10.1177/1534735419846400 · Cited by 17
Alfuraih, A. M., Tan, A. L., O’Connor, P., Emery, P., & Wakefield, R. J. (2019) · Modern Rheumatology · doi:10.1080/14397595.2019.1645374 · Cited by 15
“However, sample between 12 and 30 per group of interest has been recommended as a general rule of thumb to provide an acceptable precision around the estimates [29,30].”
BackgroundQureshi, A. I., Jahngir, M. U., Qualls, K. E., Akıncı, Y., Lobanova, I., Liaqat, J., Gao, X., Akhtar, I. N., Kraus, J., Uzun, G., et al. (2020) · Journal of Neuroimaging · doi:10.1111/jon.12714 · Cited by 15
Pincus, T., Anwar, S., McCracken, L. M., McGregor, A. H., Graham, L., Collinson, M., Farrin, A. J., & on behalf of the OBI Trial Management Team (2013) · Trials · doi:10.1186/1745-6215-14-172 · Cited by 18
“Recommendations for feasibility studies propose that the analysis dataset comprises a minimum of 30 participants for each arm in order to estimate parameters for future sample size calculations [32-34].”
BackgroundSurendran, G., Saye, J., Jalil, S. B. M., Spreadborough, J., Duong, K., Shatwan, I. M., Lilley, D., Heinrich, M., Dodd, G. F., & Surendran, S. (2025) · Frontiers in Nutrition · doi:10.3389/fnut.2025.1405796 · Cited by 18
“Furthermore, as noted by Docherty et al. (13), for the purposes of a pilot study, a sample size of between 12 and 30 per group is recommended (18, 19), therefore, considering potential drop-outs, the aim was to recruit 20 participants in total.”
Hoogerbrugge, A. J., Hooge, I. T. C., Hessels, R. S., & Strauch, C. (2025) · Behavior Research Methods · doi:10.3758/s13428-025-02754-8 · Cited by 15
“In contrast to the many guidelines and tools for determining sample size where statistical tests are involved (e.g., Browne, 1995; Adcock, 1997; Brysbaert, 2019; Faul et al., 2007), this is not straightforward for scene viewing comparisons, because the commonly used metrics (e.g., NSS, AUC) are not…”
Mitchell, F., Jahoda, A. J., Hankey, C., Matthews, L., Murray, H. M., & Melville, C. A. (2016) · Pilot and Feasibility Studies · doi:10.1186/s40814-015-0044-9 · Cited by 17
“The proposed number of participants is based on recommendations from authors specialising in feasibility and pilot studies [41, 42].”
MethodFarrand, P. A., Woodford, J., Coumoundouros, C., & Svedin, F. (2020) · The Cognitive Behaviour Therapist · doi:10.1017/s1754470x20000239 · Cited by 17
“The recommended sample size of 30 participants per arm for pilot RCTs was used (Browne, 1995), consistent with the median sample size found in pilot RCTs (Billingham et al ., 2013).”
Bethell, D., Saunders, D., Jongkaewwattana, A., Kramyu, J., Arunee, T., Wiboon-ut, S., Yongvanitchit, K., Limsalakpetch, A., Kum-Arb, U., Uthaimongkol, N., et al. (2013) · PLoS ONE · doi:10.1371/journal.pone.0059674 · Cited by 14
“This was a descriptive pilot feasibility study, designed to detect trends in safety and immune responses that could warrant expanded investigations of potentially promising combinations [13] [14].”
BackgroundXu, Z., Sanjuan, V. M., Merino‐Sanjuán, M., Merino, V., & García‐Arieta, A. (2020) · Pharmaceutics · doi:10.3390/pharmaceutics12121159 · Cited by 14
“Importantly, small pilot studies are known to underestimate the variance in clinical trials [11] and the upper boundary of the confidence interval of the variance estimation has been proposed for the sample size calculation [12].”
BackgroundMcKeown, E., McGraw, C., Holder, P., Shand, J., & Hirani, S. P. (2022) · JMIR Formative Research · doi:10.2196/36517 · Cited by 17
“For pilot studies, the sample size for quantitative components is suggested to be 30 [23,24], which will allow parameter estimates and loss to follow-up rates for subsequent large studies.”
BackgroundSavoring as an intervention for cannabis use: acceptability, feasibility, and preliminary results
Mian, M. N., & Earleywine, M. (2022) · Addiction Research & Theory · doi:10.1080/16066359.2022.2160871 · Cited by 14
“This sample size is appropriate for conducting descriptive analyses, determining a primary outcome measure for a fully developed clinical trial, and calculating effect sizes (Browne 1995).”
MethodAmano, S., Ludin, A. F. M., Clift, R., Nakazawa, M., Law, T., Rush, L. J., Manini, T. M., Thomas, J. S., Russ, D. W., & Clark, B. C. (2016) · Trials · doi:10.1186/s13063-016-1214-7 · Cited by 14
“than 95 % of the time by overestimating the population SD [70].”
BackgroundGask, L. L., Coupe, N., McElvenny, D. M., & Green, G. (2017) · British Journal of Guidance and Counselling · doi:10.1080/03069885.2017.1335391 · Cited by 14
“Lancaster, Dodd, and Williamson (2014) citing Browne (1995) state that for a pilot study (such as this) a general rule of thumb is to take 30 participants or greater to estimate a parameter, following which the pilot study should be used to calculate the sample size required for the full-scale…”
Background
Metadata
- Authors
- Browne, R. H.
- Affiliations
- Texas Scottish Rite Hospital for Children (US)
- Corresponding author
- Browne, Richard H.
- Venue
- Statistics in Medicine
- Publisher
- Wiley
- Published
- 1995-09-15
- Volume
- 14
- Issue
- 17
- Pages
- 1933-1940
- ISSN
- 0277-6715 (ISSN-L)
- 1097-0258
- OpenAlex
- W2130607817
- Cited by
- 1,007
- References
- 8
- Citations by year
- 2026: 66 · 2025: 115 · 2024: 129 · 2023: 103 · 2022: 98 · 2021: 119 · 2020: 76 · 2019: 48 · 2018: 30 · 2017: 48 · 2016: 46 · 2015: 22 · 2014: 19 · 2013: 20 · 2012: 12
- FWCI
- 1.41
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Statistics in Medicine (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Physical Sciences › Mathematics › Statistics and Probability › Probability and Statistical Research
- Keywords
- sample size determination, pilot study, t-test, statistical power, sample standard deviation, Monte Carlo simulation
- MeSH terms
- Clinical Trials as Topic, Humans, Monte Carlo Method, Pilot Projects, Reproducibility of Results, Confidence Intervals
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.