The Usefulness of the "Fail-Safe" Statistic in Meta-Analysis
The DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.
Abstract
Rosenthal (1979, 1984) proposed that a fail-safe N statistic be calculated for all meta-analytic studies to aid researchers in detecting instability in meta-analytic results. However, few meta-analysts seem to have calculated fail-safe N's, despite the ease of doing so. As research exploring the practical utility of this statistic is nonexistent, this study evaluated the utility of fail-safe N's by computing them for studies in three research domains in which discrepant conclusions were reached by initial meta-analyses (with limited samples) and by subsequent follow-ups (with more inclusive samples). The current results suggest that calculation of a fail-safe N may have led to more cautious and circumspect interpretations of previous meta-analytic results. Implications of these findings for future meta-analytic research are discussed
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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
A Fail-Safe N for Effect Size in Meta-Analysis
Orwin, R. G. (1983) · Journal of Educational Statistics · doi:10.2307/1164923 · Cited by 2,064
Statistical Power Analysis for the Behavioral Sciences
Wassertheil, S., & Cohen, J. (1970) · Biometrics · doi:10.2307/2529115 · Cited by 1,383
Wolf, F. (1986) · doi:10.4135/9781412984980 · Cited by 784
Quantitative methods for literature reviews
Green, B. F., & Hall, J. A. (1984) · Annual Review of Psychology · doi:10.1146/annurev.ps.35.020184.000345 · Cited by 213
Cited by
Maxims or myths of beauty? A meta-analytic and theoretical review
Langlois, J. H., Kalakanis, L., Rubenstein, A. J., Larson, A., Hallam, M., & Smoot, M. (2000) · Psychological Bulletin · doi:10.1037/0033-2909.126.3.390 · Cited by 2,429
A systematic review and meta-analysis of the cognitive correlates of bilingualism
Adesope, O. O., Lavin, T., Thompson, T., & Ungerleider, C. (2010) · Review of Educational Research · doi:10.3102/0034654310368803 · Cited by 1,132
Bai, S., Hew, K. F., & Huang, B. (2020) · Educational Research Review · doi:10.1016/j.edurev.2020.100322 · Cited by 994
Destination image and tourist behavioural intentions: A meta-analysis
Afshardoost, M., & Eshaghi, M. S. (2020) · Tourism Management · doi:10.1016/j.tourman.2020.104154 · Cited by 567
Insight in schizophrenia: a meta-analysis
Mintz, A. R., Dobson, K. S., & Romney, D. M. (2003) · Schizophrenia Research · doi:10.1016/s0920-9964(02)00316-x · Cited by 563
Rethinking the Use of Tests: A Meta-Analysis of Practice Testing
Adesope, O. O., Trevisan, D. A., & Sundararajan, N. (2017) · Review of Educational Research · doi:10.3102/0034654316689306 · Cited by 544
Shakespeare-Finch, J., & Lurie-Beck, J. (2014) · Journal of Anxiety Disorders · doi:10.1016/j.janxdis.2013.10.005 · Cited by 504
Cognitive reserve and cognitive function in healthy older people: a meta-analysis
Opdebeeck, C., Martyr, A., & Clare, L. (2016) · Aging, Neuropsychology, and Cognition · doi:10.1080/13825585.2015.1041450 · Cited by 465
Depression and risk of mortality in individuals with diabetes: a meta-analysis and systematic review
Park, M., Katon, W. J., & Wolf, F. M. (2013) · General Hospital Psychiatry · doi:10.1016/j.genhosppsych.2013.01.006 · Cited by 355
“The fail-safe N value is larger than the recommended 5k+10 limit (=60) [52].”
BackgroundSharpe, D. (1997) · Clinical Psychology Review · doi:10.1016/s0272-7358(97)00056-1 · Cited by 334
Residents’ impact perceptions of and attitudes towards tourism development: a meta-analysis
Gursoy, D., Ouyang, Z., Nunkoo, R., & Wei, W. (2019) · Journal of Hospitality Marketing & Management · doi:10.1080/19368623.2018.1516589 · Cited by 331
(2004) · PubMed · Cited by 280
Pellicane, M. J., & Ciesla, J. A. (2022) · Clinical Psychology Review · doi:10.1016/j.cpr.2021.102113 · Cited by 272
Do instructional interventions influence college students’ critical thinking skills? A meta-analysis
Niu, L., Behar-Horenstein, L. S., & Garvan, C. W. (2013) · Educational Research Review · doi:10.1016/j.edurev.2012.12.002 · Cited by 244
Schofield, T. J., Lee, R. D., & Merrick, M. T. (2013) · Journal of Adolescent Health · doi:10.1016/j.jadohealth.2013.05.004 · Cited by 242
“Use of a fail-safe index helps researchers to evaluate the probable stability of their results in the face of new or undiscovered evidence [23].”
BackgroundBorenstein, M. (2005) · Publication Bias in Meta‐Analysis · doi:10.1002/0470870168.ch11 · Cited by 178
Lo, C. K., & Hew, K. F. (2019) · Journal of Engineering Education · doi:10.1002/jee.20293 · Cited by 174
“Carson, Schriesheim, and Kinicki (1990) stated that “if the fail-safe N (X) is relatively small in comparison to the number of studies in the meta-analysis (k), then only tenuous conclusions should be drawn” (p. 239).”
BackgroundPreti, A., & Cella, M. (2010) · Schizophrenia Research · doi:10.1016/j.schres.2010.07.026 · Cited by 167
Publication Bias in Meta-Analysis: Confidence Intervals for Rosenthal’s Fail-Safe Number
Fragkos, K. C., Tsagris, M., & Frangos, C. C. (2014) · International Scholarly Research Notices · doi:10.1155/2014/825383 · Cited by 159
Richardson, M., McEwan, K., Maratos, F., & Sheffield, D. (2016) · Evolutionary Psychological Science · doi:10.1007/s40806-016-0065-5 · Cited by 155
“…by calculating the fail safe N (Rosenthal 1979); because fail safe N is biased towards overestimating the number of null studies required to render the overall effect size non-significant (Carson et al. 1990), a funnel plot of the standard error by the standard mean differences was generated.”
BackgroundMethodAcar, F., Seurinck, R., Eickhoff, S. B., & Moerkerke, B. (2018) · PLOS ONE · doi:10.1371/journal.pone.0208177 · Cited by 154
“The measure provides an addition to the interpretation of results because it helps in the assessment of the confidence that can be placed in them [18] by showing how stable they remain in the presence of unreported information that may differ systematically from the information that is included.”
BackgroundJónsson, H., & Hougaard, E. (2009) · Acta Psychiatrica Scandinavica · doi:10.1111/j.1600-0447.2008.01270.x · Cited by 110
“As a supplement to these analyses, Fail-Safe N was calculated as a measure of how vulnerable meta-analysis ES findings are to the possibility of undiscovered studies (17).”
MethodXia, Q., Li, W., Yang, Y., Weng, X., & Chiu, T. K. F. (2025) · Computers and Education: Artificial Intelligence · doi:10.1016/j.caeai.2025.100455 · Cited by 45
Metadata
- Authors
- Carson, K. P., Schriesheim, C. A., & Kinicki, A. J.
- Venue
- Educational and Psychological Measurement
- Publisher
- SAGE Publications
- Published
- 1990-06
- Volume
- 50
- Issue
- 2
- Pages
- 233-243
- ISSN
- 0013-1644 (ISSN-L)
- 1552-3888
- OpenAlex
- W1981129099
- Cited by
- 99
- References
- 4
- FWCI
- 1.46
- Type
- journal article
- Language
- en
- Open access
- closed
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.