Statcheck
Statcheck is a software package written in the programming language R. Its purpose is to detect statistical errors in peer-reviewed psychology articles[1] by searching papers for statistical results, redoing the calculations described in each paper, and comparing the two values to see if they match.[2] It takes advantage of the fact that psychological research papers tend to report their results in accordance with the guidelines published by the American Psychological Association (APA).[3] This leads to several disadvantages: it can only detect results reported completely and in exact accordance with the APA's guidelines,[4] and it cannot detect statistics that are only included in tables in the paper.[5] Another limitation is that Statcheck cannot deal with statistical corrections to test statistics, like Greenhouse–Geisser or Bonferroni corrections, which actually make tests more conservative.[6] Some journals have begun piloting Statcheck as part of their peer review process.
Validity
In 2017, Statcheck's developers published a preprint paper concluding that the program accurately identified statistical errors over 95% of the time.[7] This validity study comprised more than 1,000 hand-checked tests among which 5.00% turned out to be inconsistent.[8] The study found that Statcheck recognized 60% of all statistical tests. A reanalysis of these data found that if the program flagged a test as inconsistent, it was correct in 60.4% of cases. Reversely, if a test was truly inconsistent, Statcheck flagged it in an estimated 51.8% of cases (this estimate included the undetected tests and assumed that they had the same rate of inconsistencies as the detected tests). Overall, Statcheck's accuracy was 95.9%, half a percentage point higher than the chance level of 95.4% expected when all tests are simply taken at face value. Statcheck was conservatively biased (by about one standard deviation) against flagging tests.[9]
More recent research has used Statcheck on papers published in Canadian psychology journals, finding similar rates of statistical reporting errors as the original authors based on a 30-year sample of such articles. The same study also found many typographical errors in online versions of relatively old papers, and that correcting for these reduced the estimated percent of tests that were erroneously reported.[10]
History
Statcheck was first developed in 2015 by Michele Nuijten of Tilburg University and Sacha Epskamp of the University of Amsterdam.[11][7] Later that year, Nuijten and her colleagues published a paper using Statcheck on over 30,000 psychology papers and reported that "half of all published psychology papers...contained at least one p-value that was inconsistent with its test".[12] The study was subsequently written up favorably in Nature.[13][14] In 2016, Nuijten and Epskamp both received the Leamer-Rosenthal Prize for Open Social Science from the Berkeley Initiative for Transparency in the Social Sciences for creating Statcheck.[15]
In 2016, Tilburg University researcher Chris Hartgerink used Statcheck to scan over 50,000 psychology papers and posted the results to PubPeer; he subsequently published the data he extracted from these papers in an article in the journal Data.[13][16] Hartgerink told Motherboard that "We're checking how reliable is the actual science being presented by science".[17] He also told Vox that he intended to use Statcheck to perform a function similar to a spell checker software program.[11] Hartgerink's action also sent email alerts to every researcher who had authored or co-authored a paper that it had flagged. These flaggings, and their posting on a public forum, proved controversial, prompting the German Psychological Society to issue a statement condemning this use of Statcheck.[13] Psychologist Dorothy V.M. Bishop, who had two of her own papers flagged by Statcheck, criticized the program for publicly flagging many papers (including one of her own) despite not having found any statistical errors in it.[18] Other critics alleged that Statcheck had reported the presence of errors in papers that did not actually contain them, due to the tool's failure to correctly read statistics from certain papers.[19]
Journals that have begun piloting the use of Statcheck as part of their peer review process include Psychological Science,[20] the Canadian Journal of Human Sexuality,[21] and the Journal of Experimental Social Psychology.[22] The open access publisher PsychOpen has also used it on all papers accepted for publication in their journals since 2017.[23]
See also
- Abuse of statistics
- Misuse of p-values
- Metascience
References
- Nuijten, Michèle B. (2017-02-28). "BayesMed and statcheck". Aps Observer. 30 (3). Retrieved 2018-10-18.
- Baker, Monya (2016-11-25). "Stat-checking software stirs up psychology". Nature. 540 (7631): 151–152. Bibcode:2016Natur.540..151B. doi:10.1038/540151a. ISSN 0028-0836. PMID 27905454.
- Wren, Jonathan D. (2018-06-15). "Algorithmically outsourcing the detection of statistical errors and other problems". The EMBO Journal. 37 (12): e99651. doi:10.15252/embj.201899651. ISSN 0261-4189. PMC 6003655. PMID 29794111.
- Colombo, Matteo; Duev, Georgi; Nuijten, Michèle B.; Sprenger, Jan (2018-04-12). "Statistical reporting inconsistencies in experimental philosophy". PLOS ONE. 13 (4): e0194360. Bibcode:2018PLoSO..1394360C. doi:10.1371/journal.pone.0194360. ISSN 1932-6203. PMC 5896892. PMID 29649220.
- van der Zee, Tim; Anaya, Jordan; Brown, Nicholas J. L. (2017-07-10). "Statistical heartburn: an attempt to digest four pizza publications from the Cornell Food and Brand Lab". BMC Nutrition. 3 (1): 54. doi:10.1186/s40795-017-0167-x. ISSN 2055-0928. PMC 7050813. PMID 32153834.
- Schmidt, Thomas (2016). "Sources of false positives and false negatives in the Statcheck algorithm". arXiv:1610.01010 [q-bio.QM].
- Chawla, Dalmeet Singh (2017-11-28). "Controversial software is proving surprisingly accurate at spotting errors in psychology papers". Science. Retrieved 2018-10-18.
- Nuijten, Michèle B. "The validity of the tool "Statcheck" in discovering statistical reporting inconsistencies". PsyArXiv.
- Schmidt, Thomas. "Statcheck does not work: All the numbers". PsyArXiv.
- Green, Christopher D.; Abbas, Sahir; Belliveau, Arlie; Beribisky, Nataly; Davidson, Ian J.; DiGiovanni, Julian; Heidari, Crystal; Martin, Shane M.; Oosenbrug, Eric (August 2018). "Statcheck in Canada: What proportion of CPA journal articles contain errors in the reporting of p-values?". Canadian Psychology. 59 (3): 203–210. doi:10.1037/cap0000139. ISSN 1878-7304.
- Resnick, Brian (2016-09-30). "A bot crawled thousands of studies looking for simple math errors. The results are concerning". Vox. Retrieved 2018-10-18.
- Nuijten, Michèle B.; Hartgerink, Chris H. J.; van Assen, Marcel A. L. M.; Epskamp, Sacha; Wicherts, Jelte M. (2015-10-23). "The prevalence of statistical reporting errors in psychology (1985–2013)". Behavior Research Methods. 48 (4): 1205–1226. doi:10.3758/s13428-015-0664-2. ISSN 1554-3528. PMC 5101263. PMID 26497820.
- Buranyi, Stephen (2017-02-01). "The high-tech war on science fraud". The Guardian. Retrieved 2018-10-18.
- Baker, Monya (2015-10-28). "Smart software spots statistical errors in psychology papers". Nature. doi:10.1038/nature.2015.18657. ISSN 1476-4687. Retrieved 2018-10-19.
- "Michèle Nuijten". Berkeley Initiative for Transparency in the Social Sciences. 2016-12-16. Retrieved 2018-10-19.
- Hartgerink, Chris (2016-09-23). "688,112 Statistical Results: Content Mining Psychology Articles for Statistical Test Results". Data. 1 (3): 14. doi:10.3390/data1030014.
- Buranyi, Stephen (2016-09-05). "Scientists Are Worried About 'Peer Review by Algorithm'". Motherboard. Retrieved 2018-10-18.
- "Here's why more than 50,000 psychology studies are about to have PubPeer entries". Retraction Watch. 2016-09-02. Retrieved 2018-10-18.
- Stokstad, Erik (2018-09-21). "The truth squad". Science. 361 (6408): 1189–1191. Bibcode:2018Sci...361.1189S. doi:10.1126/science.361.6408.1189. ISSN 0036-8075. PMID 30237339.
- Freedman, Leonard P.; Venugopalan, Gautham; Wisman, Rosann (2017-05-02). "Reproducibility2020: Progress and priorities". F1000Research. 6: 604. doi:10.12688/f1000research.11334.1. ISSN 2046-1402. PMC 5461896. PMID 28620458.
- Sakaluk, John K.; Graham, Cynthia A. (2017-11-17). "Promoting Transparent Reporting of Conflicts of Interests and Statistical Analyses at The Journal of Sex Research". The Journal of Sex Research. 55 (1): 1–6. doi:10.1080/00224499.2017.1395387. ISSN 0022-4499. PMID 29148841.
- JESP piloting the use of statcheck. Journal of Experimental Social Psychology. Retrieved 2018-10-19.
- "PsychOpen uses Statcheck tool for quality check". PsychOpen. 2017-04-10. Retrieved 2018-10-23.