HARKing: How Badly Can Cherry-Picking and Question Trolling Produce Bias in Published Results?

Abstract

The practice of hypothesizing after results are known (HARKing) has been identified as a potential threat to the credibility of research results. We conducted simulations using input values based on comprehensive meta-analyses and reviews in applied psychology and management (e.g., strategic management studies) to determine the extent to which two forms of HARKing behaviors might plausibly bias study outcomes and to examine the determinants of the size of this effect. When HARKing involves cherry-picking, which consists of searching through data involving alternative measures or samples to find the results that offer the strongest possible support for a particular hypothesis or research question, HARKing has only a small effect on estimates of the population effect size. When HARKing involves question trolling, which consists of searching through data involving several different constructs, measures of those constructs, interventions, or relationships to find seemingly notable results worth writing about, HARKing produces substantial upward bias particularly when it is prevalent and there are many effects from which to choose. Results identify the precise circumstances under which different forms of HARKing behaviors are more or less likely to have a substantial impact on a study’s substantive conclusions and the field’s cumulative knowledge. We offer suggestions for authors, consumers of research, and reviewers and editors on how to understand, minimize, detect, and deter detrimental forms of HARKing in future research.

This is a preview of subscription content, log in to check access.

Notes

  1. 1.

    When ρ is very large, ceiling effects can limit the biases produced by HARKing. When ρ is equal to or very near 0, bias is limited because the largest effect is equally likely to be negative as it is to be positive. In addition, when ρ = 0, HARKing will produce a distribution of sample effects whose mean is not changed but whose standard deviation is inflated.

  2. 2.

    Although this method is rarely encountered in the research literature, several software packages (e.g., NCSS, JMP) include an even more aggressive option—i.e., one that evaluates all possible regression models, starting with models that include two variables and examining every possible combination of predictors until the full p-variable model is tested.

References

  1. Aguinis, H., & Vandenberg, R. J. (2014). An ounce of prevention is worth a pound of cure: Improving research quality before data collection. Annual Review of Organizational Psychology and Organizational Behavior, 1, 569–595.

    Article  Google Scholar 

  2. Aguinis, H., Werner, S., Abbott, J. L., Angert, C., Park, J. H., & Kohlhausen, D. (2010). Customer-centric science: Reporting significant research results with rigor, relevance, and practical impact in mind. Organizational Research Methods, 13, 515–539.

    Article  Google Scholar 

  3. Aguinis, H., Dalton, D. R., Bosco, F. A., Pierce, C. A., & Dalton, C. M. (2011). Meta-analytic choices and judgment calls: Implications for theory building and testing, obtained effect sizes, and scholarly impact. Journal of Management, 37, 5–38.

    Article  Google Scholar 

  4. Aguinis, H., Shapiro, D. L., Antonacopoulou, E., & Cummings, T. G. (2014). Scholarly impact: A pluralist conceptualization. Academy of Management Learning and Education, 13, 623–639.

    Article  Google Scholar 

  5. Aguinis, H., Cascio, W. F., & Ramani, R. S. (2017). Science’s reproducibility and replicability crisis: International business is not immune. Journal of International Business Studies, 48, 653–663.

    Article  Google Scholar 

  6. Aguinis, H., Ramani, R. S., & Alabduljader, N. (in press). What you see is what you get? Enhancing methodological transparency in management research. Academy of Management Annals. https://doi.org/10.5465/annals.2016.0011.

  7. Bamberger, P., & Ang, S. (2016). The quantitative discovery: What is it and how to get it published. Academy of Management Discoveries, 2, 1–6.

    Article  Google Scholar 

  8. Banks, G. C., O’Boyle, E. H., Pollack, J. M., White, C. D., Batchelor, J. H., Whelpley, C. E., …, Adkins, C. L. (2016a). Questions about questionable research practices in the field of management: A guest commentary. Journal of Management, 42, 5–20.

  9. Banks, G. C., Rogelberg, S. G., Woznyj, H. M., Landis, R. S., & Rupp, D. E. (2016b). Editorial: Evidence on questionable research practices: The good, the bad and the ugly. Journal of Business and Psychology, 31, 323–338.

    Article  Google Scholar 

  10. Bedeian, A. G., Taylor, S. G., & Miller, A. N. (2010). Management science on the credibility bubble: Cardinal sins and various misdemeanors. Academy of Management Learning & Education, 9, 715–725.

    Google Scholar 

  11. Bergh, D. D., Aguinis, H., Heavey, C., Ketchen, D. J., Boyd, B. K., Su, P., Lau, C., & Joo, H. (2016). Using meta-analytic structural equation modeling to advance strategic management research: Guidelines and an empirical illustration via the strategic leadership-performance relationship. Strategic Management Journal, 37, 477–497.

    Article  Google Scholar 

  12. Bergh, D. D., Sharp, B. M., Aguinis, H., & Li, M. (2017). Is there a credibility crisis in strategic management research? Evidence on the reproducibility of study findings. Strategic Organization, 15, 423–436.

    Article  Google Scholar 

  13. Bernerth, J., & Aguinis, H. (2016). A critical review and best-practice recommendations for control variable usage. Personnel Psychology, 69, 229–283.

    Article  Google Scholar 

  14. Bettis, R. A., Ethiraj, S., Gambardella, A., Helfat, C., & Mitchell, W. (2016). Creating repeatable cumulative knowledge in strategic management: A call for a broad and deep conversation among authors, referees, and editors. Strategic Management Journal, 37, 257–261.

    Article  Google Scholar 

  15. Bollen, K. A. (1989). Structural equations with latent variables. New York: John Wiley.

    Google Scholar 

  16. Bosco, F. A., Aguinis, H., Singh, K., Field, J. G., & Pierce, C. A. (2015). Correlational effect size benchmarks. Journal of Applied Psychology, 100, 431–449.

    Article  Google Scholar 

  17. Bosco, F. A., Aguinis, H., Field, J. G., Pierce, C. A., & Dalton, D. R. (2016). HARKing’s threat to organizational research: Evidence from primary and meta-analytic sources. Personnel Psychology, 69, 709–750.

    Article  Google Scholar 

  18. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale: Lawrence Erlbaum.

    Google Scholar 

  19. Cortina, J. M., & Landis, R. S. (2009). When small effect sizes tell a big story, and when large effect sizes don’t. In C. E. Lance & R. J. Vandenberg (Eds.), Statistical and methodological myths and urban legends: Doctrine, verity, and fable in the organizational and social sciences (pp. 287–308). New York: Routledge.

    Google Scholar 

  20. Cortina, J. M., Aguinis, H., & DeShon, R. P. (2017). Twilight of dawn or of evening? A century of research methods in the Journal of Applied Psychology. Journal of Applied Psychology, 102, 274–290.

    Article  Google Scholar 

  21. Derksen, S., & Keselman, H. J. (1992). Backward, forward and stepwise automated subset selection algorithms: Frequency of obtaining authentic and noise variables. British Journal of Mathematical and Statistical Psychology, 45, 265–282.

    Article  Google Scholar 

  22. Edwards, J. R., Berry JW. (2010). The presence of something or the absence of nothing: Increasing theoretical precision in management research. Organizational Research Methods, 13, 668–689. https://doi.org/10.1177/1094428110380467

  23. Fanelli, D. (2009). How many scientists fabricate and falsify research? A systematic review and meta-analysis of survey data. PLoS One, 4, e5738.

    Article  Google Scholar 

  24. Fisher, G., & Aguinis, H. (2017). Using theory elaboration to make theoretical advancements. Organizational Research Methods, 20, 438–464.

    Article  Google Scholar 

  25. Grand, J. A., Rogelberg, S. G., Allen, T. D., Landis, R. S., Reynolds, D. H., Scott, J. C., Tonidandel, S., & Truxillo, D. M. (in press). A systems-based approach to fostering robust science in industrial-organizational psychology. Industrial and Organizational Psychology: Perspectives on Science and Practice.

  26. Hambrick DC. (2007). The field of management’s devotion to theory: Too much of a good thing? Academy of Management Journal, 50, 1346–1352. http://doi.org/10.2307/20159476

  27. Harrell, H. (2011). Regression modeling strategies with applications to linear models, logistic regression and survival analysis. New York: Springer-Verlag.

    Google Scholar 

  28. Hayduk, L. A. (1987). Structural equation modeling with LISREL: Essentials and advances. Baltimore: Johns Hopkins University Press.

    Google Scholar 

  29. Hitchcock, C., & Sober, E. (2004). Prediction versus accommodation and the risk of overfitting. British Journal for the Philosophy of Science, 55, 1–34.

    Article  Google Scholar 

  30. Hollenbeck, J. H., & Wright, P. M. (2017). Harking, sharking, and tharking: Making the case for post hoc analysis of scientific data. Journal of Management, 43, 5–18.

    Article  Google Scholar 

  31. Honig, B., Lampel, J., Siegel, D., & Drnevich, P. (2014). Ethics in the production and dissemination of management research: Institutional failure or individual fallibility. Journal of Management Studies, 51, 118–142.

    Article  Google Scholar 

  32. Hubbard R, Armstrong JS. (1997). Publication bias against null results. Psychological Reports, 80, 337–338. https://doi.org/10.2466/pr0.1997.80.1.337

  33. Jensen, A. (1980). Bias in mental testing. New York: Free Press.

    Google Scholar 

  34. John, L. K., Loewenstein, G., & Prelec, D. (2012). Measuring the prevalence of questionable research practices with incentives for truth-telling. Psychological Science, 23, 524–532.

    Article  Google Scholar 

  35. Judd, C. M., & McClelland, G. H. (1989). Data analysis: A model comparison approach. New York: Harcourt.

    Google Scholar 

  36. Kerr, N. L. (1998). HARKing: Hypothesizing after the results are known. Personality & Social Psychology Review, 2, 196.

    Article  Google Scholar 

  37. Ketchen, D. J., Boyd, B. K., & Bergh, D. D. (2008). Research methodology in strategic management past accomplishments and future challenges. Organizational Research Methods, 11, 643–658.

    Article  Google Scholar 

  38. Ketchen, D. J., Ireland, R. D., & Baker, L. T. (2013). The use of archival proxies in strategic management studies: Castles made of sand? Organizational Research Methods, 16, 32–42.

    Article  Google Scholar 

  39. Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York: Guilford Press.

    Google Scholar 

  40. Landers, R. N., Brusso, R. C., Cavanaugh, K. J., & Collmus, A. B. (2016). A primer on theory-driven web scraping: Automatic extraction of big data from the Internet for use in psychological research. Psychological Methods, 21, 475–492.

    Article  Google Scholar 

  41. Landis, R. S., Edwards, B. D., & Cortina, J. M. (2009). On the practice of allowing correlated residuals among indicators in structural equation models. In C. E. Lance & R. J. Vandenberg (Eds.), Statistical and methodological myths and urban legends: Doctrine, verity and fable in the organizational and social sciences (pp. 193–214). New York: Routledge/Taylor & Francis Group.

    Google Scholar 

  42. Leung, K. (2011). Presenting post hoc hypotheses as a priori: Ethical and theoretical issues. Management and Organization Review, 7, 471–479.

    Article  Google Scholar 

  43. Lipton, P. (2005). Testing hypotheses: Prediction and prejudice. Science, 307, 219–221.

    Article  Google Scholar 

  44. Lo, A. W., & MacKinlay, A. C. (1990). Data-snooping biases in tests of financial asset pricing models. Review of Financial Studies, 3, 431–467.

    Article  Google Scholar 

  45. Locke, E. A. (2007). The case for inductive theory building. Journal of Management, 33, 867–890.

    Article  Google Scholar 

  46. Locke, K., Golden-Biddle, K., & Feldman, M. S. (2008). Perspective-making doubt generative: Rethinking the role of doubt in the research process. Organization Science, 19, 907–918.

    Article  Google Scholar 

  47. Murphy, K. R., & Cleveland, J. N. (1995). Understanding performance appraisal: Social, organizational and goal-oriented perspectives. Newbury Park: Sage.

    Google Scholar 

  48. Neuroskeptic. (2012). The nine circles of scientific hell. Perspectives on Psychological Science, 7, 643–644.

    Article  Google Scholar 

  49. O’Boyle, E. H., Banks, G. C., & Gonzalez-Mulé, E. (2017). The chrysalis effect: How ugly initial results metamorphosize into beautiful articles. Journal of Management, 43, NPi. https://doi.org/10.1177/0149206314527133.

  50. Orlitzky M. (2012). How can significance tests be deinstitutionalized? Organizational Research Methods, 15, 199–228. https://doi.org/10.1177/1094428111428356

  51. Pfeffer J. (2007). A modest proposal: How we might change the process and prod- uct of managerial research. Academy of Management Journal, 50, 1334–1345. https://doi.org//10.2307/20159475

  52. Pigliucci, M. (2009). The end of theory in science? EMBO Reports, 10, 534.

    Article  Google Scholar 

  53. Shaw, J. B. (2017). Advantages of starting with theory. Academy of Management Journal, 60, 819–822.

    Article  Google Scholar 

  54. Shen, W., Kiger, T. B., Davies, S. E., Rasch, R. L., Simon, K. M., & Ones, D. S. (2011). Samples in applied psychology: Over a decade of research in review. Journal of Applied Psychology, 96, 1055–1064.

    Article  Google Scholar 

  55. Sörbom, D. (1989). Model modification. Psychometrika, 54, 371–384.

    Article  Google Scholar 

  56. Thurstone, L. L. (1934). The vectors of the mind. American Psychologist, 41, 1–32.

    Google Scholar 

  57. Tonidandel, S., King, E. B., & Cortina, J. M. (Eds.). (2016). Big data at work: The data science revolution and organizational psychology. New York: Routledge.

    Google Scholar 

  58. Wasserman, R. (2013). Ethical issues and guidelines for conducting data analysis in psychological research. Ethics and Behavior, 23, 3–15.

    Article  Google Scholar 

  59. White R. (2003). The epistemic advantage of prediction over accommodation. Mind, 112, 653–683. https://doi.10.1093/mind/112.448.653

  60. Wilkinson, L., & Task Force on Statistical Inference. (1999). Statistical methods in psychology journals: Guidelines and explanations. American Psychologist, 54, 594–604.

    Article  Google Scholar 

  61. Wing, H. (1982). Statistical hazards in the determination of adverse impact with small samples. Personnel Psychology, 35, 153–162.

    Article  Google Scholar 

  62. Wright, P. M. (2016). Ensuring research integrity: An editor’s perspective. Journal of Management, 42, 1037–1043.

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Kevin R. Murphy.

Appendix. R Codes Used in Simulation Studies

Appendix. R Codes Used in Simulation Studies

The codes below calculate the expected results if 100% of studies engage in either cherry-picking or question trolling. The final estimates of the values expected if some proportion of all studies involve either cherry-picking or question trolling are obtained by calculating the weighted average (weighted by estimated prevalence) of the values produced by the codes below and the expected value of R = 0.20 if there is no cherry-picking or question trolling.

R code for cherry-picking

figurea

R code for question trolling

figureb

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Murphy, K.R., Aguinis, H. HARKing: How Badly Can Cherry-Picking and Question Trolling Produce Bias in Published Results?. J Bus Psychol 34, 1–17 (2019). https://doi.org/10.1007/s10869-017-9524-7

Download citation

Keywords

  • HARKing
  • Simulation
  • Publication bias
  • Data snooping