The DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.
Using Effect Size—or Why the P Value Is Not Enough
Open accessAbstract
These statements about the importance of effect sizes were made by two of the most influential statistician-researchers of the past half-century. Yet many submissions to Journal of Graduate Medical Education omit mention of the effect size in quantitative studies while prominently displaying the P value. In this paper, we target readers with little or no statistical background in order to encourage you to improve your comprehension of the relevance of effect size for planning, analyzing, reporting, and understanding education research studies.In medical education research studies that compare different educational interventions, effect size is the magnitude of the difference between groups. The absolute effect size is the difference between the average, or mean, outcomes in two different intervention groups. For example, if an educational intervention resulted in the improvement of subjects' examination scores by an average total of 15 of 50 questions as compared to that of another intervention, the absolute effect size is 15 questions or 3 grade levels (30%) better on the examination. Absolute effect size does not take into account the variability in scores, in that not every subject achieved the average outcome.In another example, residents' self-assessed confidence in performing a procedure improved an average of 0.4 point on a Likert-type scale ranging from 1 to 5, after simulation training. While the absolute effect size in the first example appears clear, the effect size in the second example is less apparent. Is a 0.4 change a lot or trivial? Accounting for variability in the measured improvement may aid in interpreting the magnitude of the change in the second example.Thus, effect size can refer to the raw difference between group means, or absolute effect size, as well as standardized measures of effect, which are calculated to transform the effect to an easily understood scale. Absolute effect size is useful when the variables under study have intrinsic meaning (eg, number of hours of sleep). Calculated indices of effect size are useful when the measurements have no intrinsic meaning, such as numbers on a Likert scale; when studies have used different scales so no direct comparison is possible; or when effect size is examined in the context of variability in the population under study.Calculated effect sizes can also quantitatively compare results from different studies and thus are commonly used in meta-analyses.The effect size is the main finding of a quantitative study. While a P value can inform the reader whether an effect exists, the P value will not reveal the size of the effect. In reporting and interpreting studies, both the substantive significance (effect size) and statistical significance (P value) are essential results to be reported.For this reason, effect sizes should be reported in a paper's Abstract and Results sections. In fact, an estimate of the effect size is often needed before starting the research endeavor, in order to calculate the number of subjects likely to be required to avoid a Type II, or β, error, which is the probability of concluding there is no effect when one actually exists. In other words, you must determine what number of subjects in the study will be sufficient to ensure (to a particular degree of certainty) that the study has acceptable power to support the null hypothesis. That is, if no difference is found between the groups, then this is a true finding.Statistical significance is the probability that the observed difference between two groups is due to chance. If the P value is larger than the alpha level chosen (eg, .05), any observed difference is assumed to be explained by sampling variability. With a sufficiently large sample, a statistical test will almost always demonstrate a significant difference, unless there is no effect whatsoever, that is, when the effect size is exactly zero; yet very small differences, even if significant, are often meaningless. Thus, reporting only the significant P value for an analysis is not adequate for readers to fully understand the results.For example, if a sample size is 10 000, a significant P value is likely to be found even when the difference in outcomes between groups is negligible and may not justify an expensive or time-consuming intervention over another. The level of significance by itself does not predict effect size. Unlike significance tests, effect size is independent of sample size. Statistical significance, on the other hand, depends upon both sample size and effect size. For this reason, P values are considered to be confounded because of their dependence on sample size. Sometimes a statistically significant result means only that a huge sample size was used.3A commonly cited example of this problem is the Physicians Health Study of aspirin to prevent myocardial infarction (MI).4 In more than 22 000 subjects over an average of 5 years, aspirin was associated with a reduction in MI (although not in o
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Concerning effect size, the r-value was 0.34.This would be counted as medium effect size (Sullivan & Feinn, 2012).”
From Teaching and Learning Science Using YouTube Videos and Discovery Learning in Primary School (2020) · cites this paper as Sullivan & Feinn, 2012
Experimental — accuracy not yet measuredIn the citing paper
“effect sizes, however, are independent of sample size (Sullivan and Feinn, 2012).”
From A chatbot-delivered intervention for optimizing social media use and reducing perceived isolation among rural-living LGBTQ+ youth: Development, acceptability, usability, satisfaction, and utility (2023) · cites this paper as Sullivan and Feinn, 2012
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
An effect size primer: A guide for clinicians and researchers.
Ferguson, C. J. (2009) · Professional Psychology: Research and Practice · doi:10.1037/a0015808 · Cited by 3,467
Things I have learned (so far).
Cohen, J. (1990) · American Psychologist · doi:10.1037/0003-066x.45.12.1304 · Cited by 1,781
Beyond significance testing: Reforming data analysis methods in behavioral research.
Kline, R. B. (2004) · doi:10.1037/10693-000 · Cited by 1,034
Meta-Analysis of Multiple Primary Prevention Trials of Cardiovascular Events Using Aspirin
Bartolucci, A. A., Tendera, M., & Howard, G. (2011) · The American Journal of Cardiology · doi:10.1016/j.amjcard.2011.02.325 · Cited by 185
Research works printed but not linked: 1
- Coe R. It's the effect size, stupid: what “effect size” is and why it is important. Paper presented at the 2002 Annual Conference of the British Educational Research Association, University of Exeter, Exeter, Devon, England, September 12–14, 2002. http://www.leeds.ac.uk/educol/documents/00002182.htm. Accessed March 23, 2012.
Cited by
Sorted by citations
Items 201–300 of 1,078
Phetlerthirunkul, P., Tsianakas, V., Ayis, S., & Harris, R. (2025) · International Journal of Nursing Studies · doi:10.1016/j.ijnurstu.2025.105303 · Cited by 4
Antosiak-Cyrak, K., Demuth, A., Czerniak, U., Ratajczak, J., Bryl, E., Kowalski, P., Wochna, K., Lewandowska, M., & Domaszewska, K. (2025) · Nutrients · doi:10.3390/nu17213396 · Cited by 3
“The effect sizes (Cohen’s d) were calculated as the ratio of the mean difference to the standard deviation of the difference, and categorized according to Cohen’s criteria as small ( ≥ 0.2 and 0.5), moderate ( ≥ 0.5 and 0.8), or large ( ≥ 0.8) [36,37].”
InfluentialLee, C., Quito, J., Poteat, T., Mourikes, V. E., Flaws, J. A., & Mahoney, M. M. (2025) · Toxics · doi:10.3390/toxics13110918 · Cited by 5
Shikino, K., Ozaki, N., Araki, N., Yamauchi, K., Tajima, H., Kasai, H., Shimizu, I., & Ito, S. (2025) · Journal of general and family medicine · doi:10.1002/jgf2.70072 · Cited by 4
“Effect sizes (Cohen's d for paired samples) were specified a priori and interpreted according to conventional thresholds (small = 0.20, medium = 0.50, large = 0.80) [31].”
InfluentialByrne, D., Doyle, F., Brannick, S., Carney, R. M., Cuijpers, P., Dima, A., Freedland, K., Guerin, S., Hevey, D., Kathuria, B., et al. (2025) · Psychological Medicine · doi:10.1017/s0033291725101785 · Cited by 4
“Gorter et al. (2016) only assessed mean differences and did not assess potential differences in standardized effect size outcomes, which are superior to simple mean score differences as a measure of the magnitude of observed differences (Sullivan & Feinn, 2012).”
The Topological Architecture of Brain Identity
Poetto, S., Merritt, H., Santoro, A., Rabuffo, G., Finc, K., Battaglia, D., Vaccarino, F., Saggar, M., Brovelli, A., & Petri, G. (2025) · bioRxiv · doi:10.1101/2025.06.20.660792 · Cited by 4
“To quantify the discriminability offered by different representations, we calculated the effect size [99], sep-arating the distributions of within-subject and between-subject similarity scores.”
Ludgate, S., McGowan, A., Moran, C., Attanasio, R., Žarković, M., Nagy, E. V., Negro, R., Papini, E., Cohen, C. A., Akarsu, E., et al. (2025) · Frontiers in Endocrinology · doi:10.3389/fendo.2025.1665720 · Cited by 4
“Effect size, reported alongside p-values, was used to determine the practical relevance of results, following guidelines to avoid emphasizing statistically signi fi cant but practically irrelevant fi ndings (49, 50).”
Levasseur, W., & Pozzobon, V. (2025) · Bioresource Technology · doi:10.1016/j.biortech.2025.133262 · Cited by 4
Hand, L. K., Nelson, E., Glaser, A., Russell, J., Baber, G., Gratton, M. K. P., Barry, A., Bock, K. J., & Siengsukon, C. F. (2025) · Age and Ageing · doi:10.1093/ageing/afaf287 · Cited by 4
Quantifying placebo and trial participation effects on cognitive outcome measures in aging dogs
Simon, K. E., Russell, K. J., Mondino, A., Yang, C.-C., Case, B. C., Anderson, Z., Whitley, C., Griffith, E. H., Gruen, M. E., & Olby, N. J. (2025) · GeroScience · doi:10.1007/s11357-025-01822-3 · Cited by 3
“The magnitude of these changes (i.e., the magnitude of the placebo effect or the magnitude of trial participation effect) was then quantified by calculating effect size [41].”
InfluentialKeller, J., Geier, U., & Tran, N. T. (2025) · International Journal of Molecular Sciences · doi:10.3390/ijms26157038 · Cited by 4
“Thus, we also calculated the effect size (Cohen’s d) to evaluate the magnitude of the difference [31].”
InfluentialBarranco, I., Martínez-Díaz, P., Parra, A., Martínez-Alborcia, M., Lucas, X., Rodríguez-Martínez, H., & Roca, J. (2025) · Andrology · doi:10.1111/andr.70089 · Cited by 4
Gonzalez, D. E., Melrose, D., Kukić, F., Lockie, R. G., Orr, R. M., & Dawes, J. J. (2025) · Healthcare · doi:10.3390/healthcare13131584 · Cited by 4
“For pairwise comparisons with Bonferroni post hoc corrections between groups, effect sizes were calculated using Cohen’s d and interpreted as follows: <0.2 trivial, <0.5 small, <0.8 moderate, <1.2 large, and >1.2 very large [33].”
InfluentialFreitas, C., Cooper, C., Kroch, A. E., Moineddin, R., Arbess, G., Benoit, A. C., Buchan, S. A., Chambers, C., Habanyama, M., Kendall, C., et al. (2025) · Vaccine · doi:10.1016/j.vaccine.2025.127422 · Cited by 4
Hart, C. S., & Chumanov, E. (2025) · International Journal of Sports Physical Therapy · doi:10.26603/001c.141101 · Cited by 4
Felser, S., Bonke, L. A., Glass, Ä., Strüder, D., Stolle, J., Schulze, S., Blaurock, M., Steinmetz, A., Daunheimer, J., Kriesen, U., et al. (2025) · Frontiers in Oncology · doi:10.3389/fonc.2025.1602532 · Cited by 4
Renin-angiotensin-aldosterone system activity in a female trauma-exposed sample
Sinnott, S. M., Ellis, R. A., Khawand, A., Marvar, P. J., Stein, M. B., Ressler, K. J., Huffman, J. C., & Seligowski, A. V. (2025) · Psychiatry Research · doi:10.1016/j.psychres.2025.116573 · Cited by 4
Jiao, J., Ma, S., Jiao, T., Shi, S., Gao, Y., Zhang, X., Zhao, S., & Degen, A. A. (2025) · Frontiers in Microbiology · doi:10.3389/fmicb.2025.1465078 · Cited by 4
Sebek, J., Taeprasartsit, P., Pathomparai, C., Dupuy, D. E., Wibowo, H., & Prakash, P. (2025) · Medical Physics (Lancaster) · doi:10.1002/mp.17897 · Cited by 4
Ling, J., Zahry, N. R., Chen, S., Xie, Y., Lalonde, H., Robbins, L. B., & Kerver, J. M. (2025) · Critical Public Health · doi:10.1080/09581596.2025.2505110 · Cited by 4
“…and categorical outcomes, respectively. effect size cohen’s d was calculated for continuous outcomes: 0.2 small, 0.5 medium, and 0.8 large (sullivan & Feinn, 2012). to examine outcome changes from baseline to post-intervention, we performed mixed-effects models by including time (baseline…”
InfluentialMethodological Foundations for AI-Driven Survey Question Generation
Mburu, T. K., Rong, K., McColley, C. J., & Werth, A. (2025) · arXiv.org · doi:10.48550/arxiv.2505.01150 · Cited by 4
“We also calculated Cohen’s d to measure the effect size of these differences (Sullivan & Feinn, 2012; Becker, 2000).”
InfluentialWorkman, C., & Stamatis, A. (2025) · International journal of exercise science · doi:10.70252/gbpt2217 · Cited by 3
“29 A plot was generated to visualize the difference between MT1 and MT2 scores.”
Buran, B. (2025) · Case Studies on Transport Policy · doi:10.1016/j.cstp.2025.101472 · Cited by 4
Meyers, R., Brna, M., Hogg-Cornejo, V., Donahue, C. C., Sweeney, E. A., Armento, A., & Howell, D. R. (2025) · Frontiers in Sports and Active Living · doi:10.3389/fspor.2025.1527622 · Cited by 4
Takadera, T., Thau, E., McLean, L. J., Simone, L., Schonfeld, D., Fayyaz, J., Kvatinsky, N., Litwin, S., & Harel-Sterling, M. (2026) · Resuscitation Plus · doi:10.1016/j.resplu.2026.101380 · Cited by 3
“22,23 All statistical analyses were performed using Stata MP version 19.0 (StataCorp LLC, College Station, TX, USA).”
InfluentialEstes, L., Broeckel, J., Rhoades, N., Kibler, G. L., Bivins, I. H., Liu, Y., Johnson, S., Dickerson, B. L., Gonzalez, D. E., Sowinski, R. J., et al. (2026) · Nutrients · doi:10.3390/nu18111743 · Cited by 3
“We conducted a comprehensive statistical and practical/clinical analysis with the IBM SPSS Statistics Version 31 software (IBM Corp., Armonk, NY, USA), using procedures previously described in detail [29,45–51].”
A comparison of deep multiomics profiles across ethnicity, geography, and age
Bararpour, N., Cao, J. Z., Wu, Y., Gupta, S., Hoopmann, M. R., Qin, R., Midha, M. K., Mireault, M., Juanes‐Velasco, P., Hanson, C. R., et al. (2026) · Cell · doi:10.1016/j.cell.2026.04.032 · Cited by 3
Approaches to Identifying Social Needs: The SECURE RCT
Cullen, D. L., Brown, R., McPeak, K. E., Min, J., McGovern, B. E., Dziedzic, S. M., & Fein, J. A. (2026) · PEDIATRICS · doi:10.1542/peds.2025-073800 · Cited by 3
Milani, A., Saiani, L., Gandini, S., Misurelli, E., Lacapra, S., Cafiero, C., Boselli, S., Massaro, M. A., Sponton, A., Zoppini, L., et al. (2026) · Diagnosis · doi:10.1515/dx-2025-0178 · Cited by 2
“…using Green ’ s formula [19] for multiple regression (n ≥ L/f 2 ) with 12 independent variables (age, gender, experience, education, ward type, sta ff size, organizational model, scenario characteristics) and a medium e ff ect size (f 2 =0.15) based on prior clinical reasoning research [15, 20, 21].”
InfluentialTaddio, A., Shah, V., McMurtry, C., Lang, E., MacDonald, N. E., Constantin, K. L., Uleryk, E., Abrams, E. M., Alexander, S., Bucci, L., et al. (2026) · The Clinical Journal of Pain · doi:10.1097/ajp.0000000000001372 · Cited by 3
Nikolic, S., de Andrade, M. O., Stanisavljević, Ž., Jovanović, K., Cincović, J., Radosavljević, S., & Tubbal, F. (2026) · Computer Applications in Engineering Education · doi:10.1002/cae.70178 · Cited by 3
The prevalence of Arctic multilayer clouds and their observed and modelled characteristics
Wallentin, G., Ickes, L., Achtert, P., Tesche, M., & Hoose, C. (2026) · Atmospheric Chemistry and Physics · doi:10.5194/acp-26-3069-2026 · Cited by 3
“Median and mean N ice for CMT9 and CMT5 are tabulated in Table 1.”
Haverila, K. C., Haverila, M. J., & Zarea, H. (2026) · Journal of Marketing Analytics · doi:10.1057/s41270-025-00454-y · Cited by 4
Farnana, K., Alzubi, A. B., Aljuhmani, H. Y., & Khadem, A. (2026) · Kybernetes · doi:10.1108/k-12-2024-3393 · Cited by 13
Kang, M. S., Kak, A., Osborne, L., Herring, J., & Kim, H. K. W. (2026) · The Bone & Joint Journal · doi:10.1302/0301-620x.108b2.bjj-2024-1473.r3 · Cited by 3
Elashry, M., Kassim, Y., Hu, B., Sheng, H., Xu, G., Elashry, H., & Zhang, K. (2025) · International Journal of Molecular Sciences · doi:10.3390/ijms27010397 · Cited by 3
“The magnitude of these differences was assessed using Cohen’s d effect size [13,14], computed with the effsize R package v0.”
Depression symptoms and mental health well-being among Bangladeshi nursing students
Roy, A. K., Hossain, M. I., Chowdhury, S. A., Mezan, M. A. Q., Akter, M., Akter, S., Akter, S., Banna, M. H. A., Chowdhury, S., Kabir, H., et al. (2025) · Discover Social Science and Health · doi:10.1007/s44155-025-00335-w · Cited by 3
“Cohen’s conventional thresholds are d = 0.2 (small), 0.5 (medium), 0.8 (large), and f = 0.1 (small), 0.25 (medium), 0.4 (large) [35, 36].”
InfluentialLennon, A. R., Killelea, C., Faherty, M. S., & Sell, T. C. (2025) · International Journal of Sports Physical Therapy · doi:10.26603/001c.147061 · Cited by 3
Behavioral experiments to reduce food loss and waste: A meta-analysis
Bonte, G., Compernolle, T., Jaeger, S. D., & Rousseau, S. (2025) · Food Quality and Preference · doi:10.1016/j.foodqual.2025.105809 · Cited by 5
Villa, T. H. C., Ruiz‐Vivanco, G., Porchia, L. M., Torres‐Rasgado, E., López‐Bayghen, E., & Gonzalez-Mejia, M. E. (2025) · Nutrition Research · doi:10.1016/j.nutres.2025.12.001 · Cited by 4
Measurement of renal tumor size: the impact of 2D versus 3D methods on the T-stage
Vermijs, S., Vangeneugden, J., Visschere, P. D., Backer, P. D., Decaestecker, K., Praet, C. V., & Debbaut, C. (2025) · Abdominal Radiology · doi:10.1007/s00261-025-05292-1 · Cited by 3
“To indicate the relevance of any found difference, the effect size, i.e. the magnitude of difference between groups, was calculated using Cohen’s d. Effect sizes were considered insignificant (d < 0.2), small (d ≥ 0.2 and d < 0.5), medium (d ≥ 0.5 and d < 0.8) or large (d ≥ 0.8) [20 ].”
InfluentialWilliams, J. R., Hylin, J. C., Fang, Y., & Lerner, Z. F. (2025) · Journal of Biomechanics · doi:10.1016/j.jbiomech.2025.113071 · Cited by 3
Early detection of G2SCH through machine learning analysis of physical examination metrics
Huang, X., Lin, R., Xiao, Y., Ma, Z., Xu, H., Wu, Y., & Li, X. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-02012-3 · Cited by 3
“Cohen classi fi ed effect sizes as small (d = 0.2), medium (d = 0.5), and large (d ≥ 0.8) 45 .”
InfluentialBarbry, A., Coquart, J. B., Bernier, M., Daubresse, A., Ovigneur, H., & Carton, A. (2025) · Mindfulness · doi:10.1007/s12671-025-02713-6 · Cited by 3
“The ES calculation requested by the CONSORT guidelines (refer to the supplementary material ESM_2) given the magnitude of the difference between two variables (Sullivan & Feinn, 2012), was able to highlight different levels of ES for both intra-and intergroup analyses.”
Climate change risks to future sustainable fishing using global seafood ecolabel data
Koerner, L. M., Palacios‐Abrantes, J., Novaglio, C., Blanchard, J. L., Melnychuk, M. C., Essington, T. E., Everett, J. D., Guiet, J., Harrison, C. S., Heneghan, R., et al. (2025) · Cell Reports Sustainability · doi:10.1016/j.crsus.2025.100555 · Cited by 1
Plaatjie, O. N., van Furth, A. M. V., Solomons, R., van der Kuip, M., & Mason, S. (2025) · International Journal of Infectious Diseases · doi:10.1016/j.ijid.2025.108219 · Cited by 3
Comparison of brain volumetry in patients with non-lesional epilepsy on 3 and 7 T MPRAGE
Schwarzer, J., Khadhraoui, E., Einspänner, E., Kukhlenko, O., Behme, D., Büntjen, L., Schmitt, F. C., & Müller, S. J. (2025) · Quantitative Imaging in Medicine and Surgery · doi:10.21037/qims-2025-1263 · Cited by 3
Georgieva-Tsaneva, G. N., Tsanev, Y.-A., Dechev, M., & Cheshmedzhiev, K. (2025) · Applied Sciences · doi:10.3390/app152010892 · Cited by 3
McLure, F. I., Yerdelen-Damar, S., Kaya, K., & Aydin, Z. (2025) · Journal for STEM Education Research · doi:10.1007/s41979-025-00164-1 · Cited by 3
“Significant differences between values can sometimes be observed between values that are quite small (Sullivan & Feinn, 2021).”
Gómez-López, J. C., Álvarez, J. M. R., Castillo-Secilla, D., & González, J. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130631 · Cited by 4
Chang, Y.-C., Liu, H.-W., Hsieh, T.-J., & Chang, S.-J. (2025) · Archives of gerontology and geriatrics (Print) · doi:10.1016/j.archger.2025.106055 · Cited by 3
Schizophrenia, Bipolar, or Major Depressive Disorder and Postacute Sequelae of COVID-19
Vekaria, V., Thiruvalluru, R. K., Verzani, Z., Abedian, S., Olfson, M., Patra, B. G., Xiao, Y.-Y., Salamon, K. S., Hoth, K., Blanceró, F., et al. (2025) · JAMA Network Open · doi:10.1001/jamanetworkopen.2025.40242 · Cited by 3
Job stress and its impact on daytime sleepiness among nurses in Palestine: a cross-sectional study
Zabin, L. M., Battat, M., Qaddumi, J. A. S., Ghawadra, S. F., Sakleh, I. R., & Abdullah, A. (2025) · Sleep Science and Practice · doi:10.1186/s41606-025-00139-6 · Cited by 3
“1 Arithmetic mean scores of the NSS subscales statistical significance does not always equate to clinical significance (Sullivan and Feinn 2012).”
Sousa, T., Pereira, A., Marques, D. L., Neiva, H. P., Rodríguez‐Rosell, D., Marinho, D. A., & Marques, M. C. (2025) · BMC Sports Science Medicine and Rehabilitation · doi:10.1186/s13102-025-01326-9 · Cited by 3
“The Hedge’s g effect size assessed the magnitude of differences (small: ≤ 0.20; moderate: 0.21–0.79; large: ≥ 0.80) [36].”
InfluentialAssessing the User Experience of the EU Mobile App for Cancer Prevention: Mixed Methods Study
Ahmed, F., Saletti, S. R. R., D'Souza, E., Espina, C., Ritchie, D., Barceló, A. M., Carbó, M., Cervera, P. R., Seum, T., Brenner, H., et al. (2025) · JMIR Formative Research · doi:10.2196/73844 · Cited by 3
“To validate the ANOVA findings, assumptions were checked: the Shapiro-Wilk and Q-Q plots for normality, the Levene test for homogeneity of variance, and the Cook distance ( 1) for outlier detection [23].”
Saljoqi, M., Ceccato, R., Orsini, F., Rossi, R., & Gastaldi, M. (2025) · Future Transportation · doi:10.3390/futuretransp5030124 · Cited by 3
“In such contexts, it is therefore more appropriate to interpret and discuss the results primarily in terms of effect sizes rather than p -values [62].”
Integrating Anatomical Priors Into a Causal Diffusion Model
Li, B., Peng, W., Li, M., Adeli, E., & Pohl, K. M. (2026) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2026.3688515 · Cited by 2
Machine Learning with Multitype Protected Attributes: Intersectional Fairness through RegularisationPreprint
Lee, H. M., Antonio, K., Avanzi, B., Marchi, L., & Zhou, R. H. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.08163 · Cited by 2
“Associated p -values are highly sensitive to sample size: as sample size increases, even minor dependencies can become statistically significant (Sullivan and Feinn, 2012; Lakens, 2022), which may lead to excessive regularisation and reduced model accuracy.”
MetaboGNN: predicting liver metabolic stability with graph neural networks and cross-species data
Park, J. H., Han, R., Jang, J., Kim, J., Paik, J., Heo, J., & Lee, Y. (2025) · Journal of Cheminformatics · doi:10.1186/s13321-025-01089-y · Cited by 2
“To understand the features that the GNN model considers critical for predicting the metabolic stability of molecules, we extracted edge importance scores using EdgeSHAPer [45] and analyzed their contributions using effect size (Cohen’s d ) [46] for each bond in the test set (Fig.”
Papageorgiou, V., Lampe, F., Sewell, J., Smith, C. J., Briggs, A., Ottridge, R., Apea, V., Brady, M., Lockyer, S., Asboe, D., et al. (2025) · Trials · doi:10.1186/s13063-025-08904-9 · Cited by 3
“This difference equates to an odds ratio of between 1.86 and 2, equivalent to a standardised effect size of approximately 0.35, which is between a small and a medium effect according to Cohen’s criteria [53].”
InfluentialUK Medical Cannabis Registry: A clinical outcomes analysis for insomnia
Aggarwal, A., Erridge, S., Cowley, I., Evans, L., Varadpande, M., Clarke, E., McLachlan, K., Coomber, R., Rucker, J., Weatherall, M. W., et al. (2025) · PLOS mental health. · doi:10.1371/journal.pmen.0000390 · Cited by 5
“Effect size for pairwise comparisons were reported as Cohen’s d values and interpreted as small (d ≥ 0.2), medium (d ≥ 0.5), and large (d ≥ 0.8), respectively [102].”
InfluentialLuan, S.-Y., Fitzgibbin-Colllins, L. K., Kohli, S., Durand, N., Hunter, S., Fleet, J. L., Viana, R., Teasell, R., & Peters, S. (2025) · NeuroRehabilitation (Reading, MA) · doi:10.1177/10538135251370599 · Cited by 3
“In the current literature, effect sizes are classi fi ed as small (d = 0.2), medium (d = 0.5), and large (d ≥ 0.8) (Sullivan & Feinn, 2012).”
InfluentialHenry, B. F., Kreager, D., Gray, J., Brant, K., Zajac, G., Lipscomb, D., Brothers, S., Schaefer, D. R., Bardele, N., & Hazelwood, A. (2025) · Health & Justice · doi:10.1186/s40352-025-00362-4 · Cited by 3
“These key programmatic outcomes will be compared between the two groups using t-tests, with effect sizes calculated using the standardized difference between group means and pooled sample standard deviations (Sullivan & Feinn, 2012).”
A Systematic Study on Prenatal Opioid Exposure: A Review of Neurodevelopmental Consequences
Cao, Q., & Kable, J. (2025) · Journal of Developmental and Behavioral Pediatrics · doi:10.1097/dbp.0000000000001407 · Cited by 3
Restoring cytonuclear harmony: Distinct strategies in Arabidopsis auto‐ and allopolyploids
Shahbazi, M., Kneřová, J., Kubíková, D., Doležalová, A., Szecówka, M., Santos, Y. D., Wendel, J. F., Sharbrough, J., & Kopecký, D. (2025) · The Plant Journal · doi:10.1111/tpj.70451 · Cited by 3
“To determine the magnitude of the difference between the groups, the effect size was calculated using Cohen’s d test with classification as follows: negligible effect d < 0.2; small effect d = 0.2 – 0.5; moderate effect d = 0.5 – 0.8; large effect d ≥ 0.8 (Sullivan & Feinn, 2012).”
InfluentialMoggio, L., Mercurio, M., Marotta, N., Longo, U. G., Gasparini, G., Ammendolia, A., & Alessandro de, S. I. R. E. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14155376 · Cited by 5
“As Sullivan and Feinn [37] emphasised, including effect size measurements is critical in quantifying the magnitude of the observed effects and assessing their practical relevance.”
Urban, S. J. (2025) · The Prison Journal · doi:10.1177/00328855251355500 · Cited by 3
“Given the small sample size, a p -value of .10 was used to identify emerging trends (Carlin et al., 2008; Sullivan & Feinn, 2012).”
Ghavimi, A., Bonerath, A., & Haunert, J. (2025) · Applied spatial analysis and policy · doi:10.1007/s12061-025-09690-2 · Cited by 3
Didier, H. A., Cappellari, A. M., Berardino, F. D., Donelli, M., Fan, X. X., Magnoni, L., Lilli, G., Barozzi, S., Zanetti, D., Arnone, F., et al. (2025) · Clinical and Experimental Dental Research · doi:10.1002/cre2.70172 · Cited by 3
“The effect size (Cohen's d = 0.82) indicates a large magnitude of difference, suggesting that the presence of tinnitus is mean-ingfully associated with increased muscular abnormality in patients with TMD (Sullivan and Feinn 2012).”
InfluentialMcCoy, L., Markert, R. J., Thoms, E., Noall, O., & Burtson, K. M. (2025) · PLoS ONE · doi:10.1371/journal.pone.0327655 · Cited by 3
Yang, Y., Huang, L.-J., Zhao, G., Xia, J., Tian, X., Liu, C., & Xia, Q. (2025) · Journal of Orthopaedic Surgery and Research · doi:10.1186/s13018-025-06050-1 · Cited by 3
“Effect size: Partial eta-squared (η 2 ) with interpretation: η 2 ≥ 0.01 (small), ≥ 0.06 (medium), ≥ 0.14 (large) [21].”
InfluentialReifenrath, J., Gardner, B., Gigler, A., Liesche-Starnecker, F., Eldershaw, S., Stone, N., & Schlegel, J. (2025) · Scientific Reports · doi:10.1038/s41598-025-08425-0 · Cited by 3
de Bot, R. D., Stevens, J., Staal, H. M., Meijer, K., & Witlox, A. (2025) · Biomechanics · doi:10.3390/biomechanics5030046 · Cited by 3
Fitzhardinge, M., Blackman, L. C., & Pilkington, P. (2025) · Clinical Psychology and Psychotherapy · doi:10.1002/cpp.70115 · Cited by 3
Tangrodchanapong, T., Jiso, A., Changkaew, P., Khachonkham, S., Thongsuk, A., Chodchavanchai, T., Khemawoot, P., Ruangsawasdi, N., Dharmasaroja, P., & Songsaad, A. (2025) · PLoS ONE · doi:10.1371/journal.pone.0325480 · Cited by 3
“In addition, based on Cohen’s formula [22], the d value of AA at 20 μM was 1.37, interpreting that the very large magnitude was observed in terms of its effect size.”
InfluentialKendall, M., Zander, T., Wolansky, R., Teixeira, L., & Kuo, P. C. (2025) · The American surgeon · doi:10.1177/00031348251331293 · Cited by 3
“17 Confounders with a small SMD should be excluded from the PSM model to avoid introducing further imbalance.”
Clement, N. D., Ajekigbe, B., Ramaskandhan, J. R., Galloway, S., Smith, K. M., Weir, D. J., & Deehan, D. J. (2025) · Bone and Joint Research · doi:10.1302/2046-3758.146.bjr-2024-0239.r1 · Cited by 4
Salman, A., Naeem, U., Ghazanfar, S., Jawed, A., & Farooq, M. (2025) · Neuro-Oncology Advances · doi:10.1093/noajnl/vdaf073 · Cited by 3
“42 Accordingly, our findings suggest that for HVLT-Total re-call, treatment with HS-WBRT showed a moderately clinically significant reduction in overall cognitive decline at six months.”
Bilau, I., Koo, B., Fu, E., Chau, W., Kwon, H., & Yang, E. (2025) · medRxiv · doi:10.1101/2025.05.21.25328033 · Cited by 3
“Ninety-five percent confidence intervals were reported for both effect sizes and mean differences to supplement p-values, following recommendations to prioritize estimation over significance testing in the context of pilot studies (Sullivan & Feinn, 2012).”
Wickie, A., Esch, N. V., & Scheerer, N. (2025) · Frontiers in Psychology · doi:10.3389/fpsyg.2025.1560100 · Cited by 3
“As such, anything above this value should indicate a meaningful effect (Cohen, 1994; Lang et al., 1998; Sullivan and Feinn, 2012; Scheerer et al., 2024a; Scheerer et al., 2024b; Scheerer et al., 2024c; Tomczak and Tomczak, 2014).”
Leopold, S., Gebhardt, M. C., Gioe, T. J., Manner, P. A., Porcher, R., Rimnac, C. M., & Wongworawat, M. (2025) · Clinical Orthopaedics and Related Research · doi:10.1097/corr.0000000000003545 · Cited by 3
Student perceptions of artificial intelligence in Doctor of Physical Therapy Education
Reoli, R., Marchese, V., Duggal, A. P., & Kaplan, K. (2025) · Physiotherapy Theory and Practice · doi:10.1080/09593985.2025.2496978 · Cited by 3
“…responses were tallied and Wilcoxon signed-rank tests were used to evaluate survey responses pre/post intervention and effect sizes, r=z/sqrt(n), were calculated with r 0.3 indicating small, 0.3≤ r < 0.5 indicating moderate, and r ≥ 0.5 indicating large effect sizes (Sullivan and Feinn, 2012).”
InfluentialBaek, D., Kim, H., Wei, Q., Lee, S.-J., & Lim, L. (2025) · Building and Environment · doi:10.1016/j.buildenv.2025.113182 · Cited by 3
Ashraf, N., Sangalli, L., Seagroves, J., & Sawicki, C. M. (2025) · Dental journal · doi:10.3390/dj13050191 · Cited by 3
“Effect sizes were computed with Cohen’s d [13].”
Kim, J. (2026) · Plastic and Reconstructive Surgery, Global Open · doi:10.1097/gox.0000000000007555 · Cited by 2
Tsai, H., Huang, C.-J., Wang, H.-C., Luo, S.-D., Rau, K.-M., & Chou, P. (2026) · Integrative Cancer Therapies · doi:10.1177/15347354261463550 · Cited by 2
LLM ‐Assisted Topic Modelling for Hate Speech Characterization
López, A. B., Pastor-Galindo, J., & Ruipérez‐Valiente, J. A. (2026) · Expert Systems · doi:10.1111/exsy.70293
NAD+ supply and redox state limit developmental speed in the Drosophila eye
veits, N., Guo, Y., He, J., Mazouni, K., Nemazanyy, I., Bres, M., picciotto, C., Mestdagh, C., Yan, Y., & Schweisguth, F. (2026) · The EMBO Journal · doi:10.1038/s44318-026-00801-4 · Cited by 3
“Differences in pathway scores were considered signi fi cant for a Cohen ’ s d effect size (noted s herein) superior to 0.2 (Sullivan and Feinn, 2012).”
InfluentialQi, L., & Zhao, L. (2026) · PLoS ONE · doi:10.1371/journal.pone.0348441 · Cited by 2
Rooney, K., Mistry, S., Pramanik, A., & Basak, A. K. (2026) · Industries · doi:10.3390/industries1010002 · Cited by 2
“The experimental dataset used to validate the machine learning framework was generated as part of a separate study on recycled epoxy-acrylate composites for DLP additive manufacturing, the full methodology and results of which have been reported in Rooney et al. [59].”
InfluentialMohtasham, F., Nazari, S. S. H., Pourhoseingholi, M. A., Kavousi, K., & Zali, M. R. (2026) · PLoS ONE · doi:10.1371/journal.pone.0341198 · Cited by 4
“Statistical comparisons employed: • Wilcoxon signed-rank test was applied for accuracy differences, • Effect size ( r = z √ N ) [65] interpreted per Cohen’s criteria (large = 0.5, medium = 0.3, small = 0.1) [66], • Holm correction to control the family-wise error rate [67], • McNemar’s test to…”
InfluentialTribhuvan, K. U., N., M., Gidhi, A., Singh, D., Kumar, S., Kumar, S., & Singh, B. K. (2026) · BMC Genomics · doi:10.1186/s12864-026-12857-x · Cited by 2
“The combination of a non-significant p -value (0.085) and a large effect size (Cohen’s d ≈ 1.02) is likely attributable to limited statistical power rather than a lack of biological effect, as effect size more accurately reflects the magnitude of response in studies with small sample sizes [45, 46].”
InfluentialNeuronal silence as a predictive biomarker and target for epileptic seizures suppression
Souza, D. L. M., Bentivoglio, L. E., Gabrick, E. C., Protachevicz, P. R., Caldas, I. L., Iarosz, K. C., Durá-Bernal, S., Batista, A. M., & Borges, F. S. (2026) · Scientific Reports · doi:10.1038/s41598-026-44063-w · Cited by 2
“These underscores that statistical significance does not necessarily imply biological relevance 60 and that increasing the stimulation amplitude does not substantially shorten up states.”
Tetlow, E. R., Wentlent, L. A., Carney, P. C., Byron, M. J., Fiani, N., Wright, A. L., Chrostek, E., Drozd, M. E., Winokur, C. E., Davis, E. M., et al. (2026) · American Journal of Veterinary Research · doi:10.2460/ajvr.26.02.0052 · Cited by 2
Santander-Rengifo, F., Carreras-Presas, C., & Cayo-Rojas, C. (2026) · Scientific Reports · doi:10.1038/s41598-026-47697-y · Cited by 2
“Overall, the present findings support a formulation-dependent interpretation of bond performance, consistent with the large effect sizes observed and with previous evidence showing that in vitro outcomes remain sensitive to protocol and sample characteristics 22,33,34 .”
Hu, X.-M., Zeng, Z.-W., Ho, C.-Y., Fong, D. T. P., & Yang, Y.-J. (2026) · Sports health · doi:10.1177/19417381261431342 · Cited by 2
“79 Given the increasing number of studies that reported bilateral deficits in people with unilateral CAI, 31 comparisons of the uninjured side versus healthy controls were also conducted.”
Chen, Y.-L., Chou, T.-Y., Chang, E., & Huang, Y. (2026) · Journal of Human Kinetics · doi:10.5114/jhk/207104 · Cited by 1
“Effect sizes were classified as small ( d = 0.2), medium ( d = 0.5), and large ( d ≥ 0.8) (Sullivan and Feinn, 2012).”
InfluentialFares‐Otero, N. E., Furukawa, Y., Sijbrandij, M., Leucht, S., Vieta, E., Cuijpers, P. J., Harrer, M., & Seedat, S. (2026) · European Neuropsychopharmacology · doi:10.1016/j.euroneuro.2026.112802 · Cited by 2
Fenner, N., Hayward, E., Bovard, B. D., Creer, S., Dunn, C., Freeman, C., & Milner, N. J. (2026) · Frontiers in Environmental Science · doi:10.3389/fenvs.2026.1729512 · Cited by 3
Cen, C., Guo, H., Chew, L. W., & Wong, N. H. (2026) · Energy and Buildings · doi:10.1016/j.enbuild.2026.117365 · Cited by 2
Metadata
- Authors
- Sullivan, G. M., & Feinn, R.
- Affiliations
- University of Connecticut (US)
- Corresponding author
- Sullivan, Gail M.
- Venue
- Journal of Graduate Medical Education
- Published
- 2012-09
- Volume
- 4
- Issue
- 3
- Pages
- 279-282
- ISSN
- 1949-8349 (ISSN-L)
- 1949-8357
- PMC
- PMC3444174
- OpenAlex
- W2107026277
- Cited by
- 5,553
- References
- 6
- Citations by year
- 2026: 607 · 2025: 877 · 2024: 759 · 2023: 547 · 2022: 574 · 2021: 499 · 2020: 416 · 2019: 419 · 2018: 320 · 2017: 222 · 2016: 215 · 2015: 83 · 2014: 9 · 2013: 4 · 2012: 2
- FWCI
- 35.75
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- Journal of Graduate Medical Education (publishedVersion)
- Europe PMC (PubMed Central) (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Social Sciences › Education › Educational Research and Methods
- Keywords
- effect size, p-values, statistical significance, substantive significance, standardized mean difference, medical education research, educational intervention, meta-analysis, simulation-based training, resident confidence, test scores, interpretation of study results, graduate medical education, reporting guidelines
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