Recovering data from summary statistics: Sample Parameter Reconstruction via Iterative TEchniques (SPRITE)
Open accessThe DOI matches its Crossref record; 12 of the 13 references it cites were found in a registry or an index.
Abstract
Scientific publications have not traditionally been accompanied by data, either during the peer review process or when published. Concern has arisen that the literature in many fields may contain inaccuracies or errors that cannot be detected without inspecting the original data. Here, we introduce SPRITE (Sample Parameter Reconstruction via Interative TEchniques), a heuristic method for reconstructing plausible samples from descriptive statistics of granular data, allowing reviewers, editors, readers, and future researchers to gain insights into the possible distributions of item values in the original data set. This paper presents the principles of operation of SPRITE, as well as worked examples of its practical use for error detection in real published work. Full source code for three software implementations of SPRITE (in MATLAB, R, and Python) and two web-based implementations requiring no local installation (1, 2) are available for readers.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
R: A Language and Environment for Statistical Computing
R Core Team (2025) · doi:10.32614/r.manuals · Cited by 352,181
The poor availability of psychological research data for reanalysis.
Wicherts, J. M., Borsboom, D., Kats, J., & Molenaar, D. (2006) · American Psychologist · doi:10.1037/0003-066x.61.7.726 · Cited by 543 · cited 1×
“For example, Wicherts et al. (2006) and Bakker and Wicherts (2011) reported receiving data from 38 out of 141 (23”
BackgroundMindless Eating: Why We Eat More Than We Think
Wansink, B. C. (2006) · Cited by 510
Bottomless Bowls: Why Visual Cues of Portion Size May Influence Intake**
Wansink, B., Painter, J. E., & North, J. (2005) · Obesity Research · doi:10.1038/oby.2005.12 · Cited by 501
The Availability of Research Data Declines Rapidly with Article Age
Vines, T. H., Albert, A. Y. K., Andrew, R. L., Débarre, F., Bock, D. G., Franklin, M. T., Gilbert, K. J., Moore, J.-S., Renaut, S., & Rennison, D. J. (2014) · Current Biology · doi:10.1016/j.cub.2013.11.014 · Cited by 482
An empirical analysis of journal policy effectiveness for computational reproducibility
Stodden, V., Seiler, J., & Ma, Z. (2018) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1708290115 · Cited by 369 · cited 4×
“However, given the high prevalence of reported statistical inconsistencies (e.g., Bakker and Wicherts, 2011) and the low prevalence of open or accessible data (Stodden et al., 2018), there is a pressing need for a tool which can “open the box” of summary statistics reported in scientific papers.”
BackgroundInfluentialWolins, L. (1962) · American Psychologist · doi:10.1037/h0038819 · Cited by 226
Just Post It: The Lesson From Two Cases of Fabricated Data Detected by Statistics Alone
Simonsohn, U. (2013) · Psychological Science · doi:10.1177/0956797613480366 · Cited by 216
Journal Data Sharing Policies and Statistical Reporting Inconsistencies in Psychology
Nuijten, M. B., Borghuis, J., Veldkamp, C. L. S., Dominguez-Alvarez, L., van Assen, M. A. L. M., & Wicherts, J. M. (2017) · Collabra: Psychology · doi:10.1525/collabra.102 · Cited by 79 · cited 3×
“However, a variety of investigative techniques exist that are able to detect anomalies in published research via the analysis of expected/observed distributions, means, standard deviations, F, t, and p values, and the relationships between these elements (e.g., Brown and Heathers, 2017; Carlisle, 2012, 2017; Nuijten e…”
BackgroundStatistical Reporting Errors and Collaboration on Statistical Analyses in Psychological Science
Veldkamp, C. L. S., Nuijten, M. B., Dominguez-Alvarez, L., van Assen, M. A. L. M., & Wicherts, J. M. (2014) · PLoS ONE · doi:10.1371/journal.pone.0114876 · Cited by 69
Cited by
Replicability, Robustness, and Reproducibility in Psychological Science
Nosek, B. A., Hardwicke, T. E., Moshontz, H., Allard, A., Corker, K. S., Dreber, A., Fidler, F., Hilgard, J., Kline Struhl, M., Nuijten, M. B., et al. (2022) · Annual Review of Psychology · doi:10.1146/annurev-psych-020821-114157 · Cited by 770
The Replication Crisis Has Led to Positive Structural, Procedural, and Community Changes
Korbmacher, M., Azevedo, F., Pennington, C. R., Hartmann, H., Pownall, M., Schmidt, K., Elsherif, M., Breznau, N., Robertson, O., Kalandadze, T., et al. (2023) · Communications Psychology · doi:10.1038/s44271-023-00003-2 · Cited by 137
“Another is the Sample Parameter Reconstruction via Iterative TEchniques (SPRITE), which reconstructs samples and estimates of the item value distributions based on reported descriptive statistics 117 .”
MethodA Guide for Calculating Study-Level Statistical Power for Meta-Analyses
Quintana, D. S. (2023) · Advances in Methods and Practices in Psychological Science · doi:10.1177/25152459221147260 · Cited by 95
“Recently developed tools can evaluate the presence of reporting errors, such as GRIM (Brown & Heathers, 2017), SPRITE (Heathers et al., 2018), and statcheck (Nuijten & Polanin, 2020).”
Responsible Research Assessment II: A specific proposal for hiring and promotion in psychologyPreprint
Gärtner, A., Leising, D., Freyer, N., Musfeld, P., Lange, J., & Schönbrodt, F. D. (2022) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/5yexm · Cited by 42
Munn, Z., Barker, T., Stern, C., Pollock, D., Ross-White, A., Klugar, M., Wiechula, R., Aromataris, E., & Shamseer, L. (2021) · JBI Evidence Synthesis · doi:10.11124/jbies-21-00138 · Cited by 44
Assessing Research Misconduct in Randomized Controlled Trials
Li, W., Bordewijk, E. M., & Mol, B. W. (2021) · Obstetrics & Gynecology · doi:10.1097/aog.0000000000004513 · Cited by 37
Teaching Good Research Practices: Protocol of a Research Master Course
Sarafoglou, A., Hoogeveen, S., Matzke, D., & Wagenmakers, E. (2019) · Psychology Learning & Teaching · doi:10.1177/1475725719858807 · Cited by 38
The Individual Participant Data Integrity Tool for assessing the integrity of randomised trials
Hunter, K. E., Aberoumand, M., Libesman, S., Sotiropoulos, J. X., Williams, J. G., Aagerup, J., Wang, R., Mol, B. W., Li, W., Barba, A., et al. (2024) · Research Synthesis Methods · doi:10.1002/jrsm.1738 · Cited by 28
Wilkinson, J., Heal, C., Antoniou, G. A., Flemyng, E., Ahnström, L., Alteri, A., Avenell, A., Barker, T. H., Borg, D. N., Brown, N. J. L., et al. (2025) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2025.111824 · Cited by 28
Self-citationBarker, T. H., Pollock, D., Stone, J., Klugar, M., Scott, A. M., Stern, C., Wiechula, R., Shamseer, L., Aromataris, E., Ross‐White, A., et al. (2023) · Research Synthesis Methods · doi:10.1002/jrsm.1613 · Cited by 26
Reckoning With Our Crisis: An Agenda for the Field of Social and Personality Psychology
Schiavone, S. R., & Vazire, S. (2022) · Perspectives on Psychological Science · doi:10.1177/17456916221101060 · Cited by 25
Detection of data fabrication using statistical tools
Hartgerink, C. H. J., Voelkel, J. G., Wicherts, J. M., & van Assen, M. A. L. M. (2019) · doi:10.31234/osf.io/jkws4 · Cited by 16
Enhancing and Accelerating Social Science Via Automation: Challenges and Opportunities
Yarkoni, T., Eckles, D., Heathers, J. A. J., Levenstein, M. C., Smaldino, P. E., & Lane, J. I. (2021) · Harvard Data Science Review · doi:10.1162/99608f92.df2262f5 · Cited by 26
Self-citationValdez, D., Vorland, C. J., Brown, A. W., Mayo-Wilson, E., Otten, J. M., Ball, R. J., Grant, S., Levy, R., Svetina, D., & Allison, D. B. (2020) · F1000Research · doi:10.12688/f1000research.26594.1 · Cited by 16
“Other examples require statistical ratios (Georgescu & Wren, 2018), or integer-based data and sample sizes (e.g., Brown & Heathers, 2017; Heathers et al., 2018), which are both challenging to automate and not recurrent across all fields.”
BackgroundWilkinson, J., Heal, C., Antoniou, G. A., Flemyng, E., Avenell, A., Barbour, V., Bordewijk, E. M., Brown, N. J. L., Clarke, M., Dumville, J., et al. (2024) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2024.111512 · Cited by 11
Self-citationAssessing the scientific integrity of the collected work of one author or author group
Nielsen, J., Bordewijk, E. M., Gurrin, L. C., Shivantha, S., Flanagan, M., Liu, S., Linn, M. M., Zhou, K. X., van Eekelen, R., Brown, N. J. L., et al. (2025) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2024.111603 · Cited by 10
Self-citationBarker, T. H., McBride, G. M., Ross‐White, A., Pollock, D., Stern, C., Hasanoff, S., Kanukula, R., Dias, M. M., Scott, A., Aromataris, E., et al. (2024) · JBI Evidence Synthesis · doi:10.11124/jbies-24-00167 · Cited by 10
Sakaluk, J. K. (2020) · Archives of Sexual Behavior · doi:10.1007/s10508-020-01795-8 · Cited by 8
Barnett, A. (2022) · F1000Research · doi:10.12688/f1000research.123002.1 · Cited by 7
“A number of automated algorithms have been created to detect numerical problems in papers, including statcheck for p-values(42) and SPRITE for summary statistics.(43) An automated check of p-values and confidence intervals found up to 5% had large errors, suggesting there are likely tens of thousands of published pape…”
MethodSchumm, W. R., & Crawford, D. W. (2019) · The Linacre Quarterly · doi:10.1177/0024363919884799 · Cited by 10
“to use SPSS to obtain a Bayesian statistic for the oldest group of children regarding self-worth, but both the means and standard deviations failed tests for legitimacy (Heathers et al. 2018), preventing us from reconstructing the data, so we could perform a Bayesian analysis. It could be argued that if part of the da…”
MethodInfluentialUsing Statistics from Binary Variables to Detect Data Anomalies, Even Possibly Fraudulent Research
Schumm, W. R., Crawford, D. W., & Lockett, L. (2019) · Psychology Research and Applications · doi:10.22606/pra.2019.14004 · Cited by 9
Research anomalies in criminology: How serious? How extensive over time? And who was responsible?
Schumm, W. R., Crawford, D. W., Lockett, L., AlRashed, A., & bin Ateeq, A. (2023) · Accountability in Research · doi:10.1080/08989621.2023.2241127 · Cited by 5
“Some approaches to detecting potentially unreliable research may not work for large samples (Brown and Heathers 2017; Heathers et al. 2018; Schumm et al. 2021).”
Data Detective Methods for Revealing Questionable Research Practices
Francis, G., & Thunell, E. (2022) · Springer eBooks · doi:10.1007/978-3-031-04968-2_6 · Cited by 4
Berrío, J. P., & Kalliokoski, O. (2025) · FEBS Letters · doi:10.1002/1873-3468.70077 · Cited by 7
“We have not made any attempts in the pre-sent investigation to check statistics [33] or tabulated numbers [34] for internal consistency, searched for pla-giarized snippets of text [35], or employed any of a number of other methods [36] proposed for detecting problematic reports.”
Les pratiques de recherche ouvertes en psychologie
Lantian, A. (2020) · Psychologie Française · doi:10.1016/j.psfr.2020.09.001 · Cited by 6
Contributions towards understanding and building sustainable sciencePreprint
Hartgerink, C. H. J. (2019) · Thesis Commons (OSF Preprints) · doi:10.31237/osf.io/4wtpc · Cited by 3
“SPRITE (Heathers et al. 2018), GRIM (Brown and Heathers 2016), and GRIMMER (Anaya 2016) are some examples of other statistical methods that test for problematic or fabricated summary statistics (see also Buyse et al.”
MethodRounded Input Variables, Exact Test Statistics (RIVETS)Preprint
Brown, N. J. L., & Heathers, J. A. J. (2019) · doi:10.31234/osf.io/ctu9z · Cited by 3
Self-citationCrone, G., & Green, C. D. (2025) · Theory & Psychology · doi:10.1177/09593543241311861 · Cited by 3
“In one study Heathers et al. (2018) investigated, they noticed a particularly strange mean/SD pair in describing the number of carrots that a group of 3–5-year-old children consumed.”
InfluentialNine Ways to Detect Possible Scientific Misconduct in Research with Small (N < 200) Samples
Schumm, W. R., Crawford, D. W., Lockett, L., AlRashed, A., & bin Ateeq, A. (2021) · Psychology Research and Applications · doi:10.22606/pra.2021.32001 · Cited by 3
Brown, N. (2019) · University of Groningen research database (University of Groningen / Centre for Information Technology) · doi:10.33612/diss.99196913 · Cited by 2
Self-citationAutomated detection of over- and under-dispersion in baseline tables in randomised controlled trialsPreprint
Barnett, A. G. (2022) · Research Square · doi:10.21203/rs.3.rs-1496802/v1 · Cited by 2
Under my umbrella: Rating scales obscure statistical power and effect size heterogeneity
Fuenderich, J., Beinhauer, L. J., & Renkewitz, F. (2025) · Behavior Research Methods · doi:10.3758/s13428-025-02879-w · Cited by 1
“Umbrella plots: Formalizing the dependency between means and standard deviations Heathers et al. (2018) visualize the possible combinations of means and standard deviations for data from rating scales in a scatter plot.”
InfluentialImprobable data characteristics and extreme effects of apple cider vinegar on weight loss
Malbouby, V., Trexler, E. T., & Heathers, J. (2025) · BMJ Nutrition Prevention & Health · doi:10.1136/bmjnph-2024-000997 · Cited by 1
Self-citationCan Social Desirability Be Detected Statistically at the Macro-Sociological Level?
Schumm, W. R., & Crawford, D. W. (2021) · Psychology Research and Applications · doi:10.22606/pra.2021.31002 · Cited by 1
“…though using 134 as a denominator does not permit an integer value for the numerator that will yield 8.8%, we used the reported value as the best data we had, even though it failed the GRIM test (Heathers et al., 2018; however, 11/125 = 8.80 if the sample size were 125 rather than 134 as reported).”
MethodInfluentialImproving Research in Educational Psychology, Psychology, and the Social SciencesPreprint
Smith, K. N., Makel, M. C., & Plucker, J. A. (2019) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/ams9e · Cited by 1
rsprite2: Identify Distributions that Match Reported Sample Parameters (SPRITE)
Wallrich, L. (2021) · doi:10.32614/cran.package.rsprite2 · Cited by 1
A note on the alpha-level: Not all alphas can be justifiedPreprint
Albers, C. J. (2019) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/erwvk · Cited by 1
Doing Science: Methods and PhilosophyPreprint
van Dongen, N. N. N. (2021) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/yegvs
「学生が将来平和を作り出すためのリーダーシップを発揮するために, ブレズレン派の大学がなすべきこととは」 (米国マンチェスター大学教授 グラディス・E・ミュアー)
徹, 片. (2014) · 北星学園大学文学部北星論集
Additional Notes of Caution: A Reply to Sakaluk (2020)
Wisman, A., & Shrira, I. (2022) · Archives of Sexual Behavior · doi:10.1007/s10508-021-02239-7
“333333), there is no correction method that enables the SPRITE test to determine the individual integers within that composite (see Heathers et al., 2018, for a discussion of this limitation). Therefore, unlike the GRIM test, it was not possible to simply adjust the sample size and re-run SPRITE tests on our descripti…”
MethodInfluentialLes pratiques de recherche ouvertes en psychologiePreprint
Lantian, A. (2020) · doi:10.31234/osf.io/tdy62
[PhD Dissertation] Good Research PracticesPreprint
Sarafoglou, A. (2022) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/7aj5h
Metadata
- Authors
- Heathers, J. A., Anaya, J., van der Zee, T., & Brown, N. J.
- Affiliations
- Northeastern University (US)
- Omnes Res, Charlottesville, Virginia, United States
- Leiden University (NL)
- University Medical Center Groningen (NL)
- Venue
- PeerJ Preprints
- Publisher
- PeerJ
- Published
- 2018-05-30
- OpenAlex
- W2807389295
- Semantic Scholar
- 580d1d571922fdb495ef4aa8fbb22dd1a373ba6c
- Cited by
- 50
- References
- 13
- Citations by year
- 2025: 12 · 2024: 5 · 2023: 4 · 2022: 9 · 2021: 7 · 2020: 3 · 2019: 9
- Type
- posted content
- Language
- en
- Open access
- green
- doi.org
- Topics
- Physical Sciences › Computer Science › Information Systems › Research Data Management Practices
- Physical Sciences › Mathematics › Statistics and Probability › Benford’s Law and Fraud Detection
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Data Visualization and Analytics
- Keywords
- sprites, reproducibility, granular data, peer review
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.