Skip to content


Below, I outline some of my main research interests, which span across the areas of psychometrics, machine learning, statistical mediation, and the evaluation of assessments.

Intersection of psychometrics and machine learning

My research investigates and develops novel uses of machine learning algorithms for our psychometric toolbox for assessment, especially in the areas of scale development and evaluation. Machine learning is a subfield of artificial intelligence that deals with data-driven, statistical algorithms that learn patterns from data and that improve through experience. Recognizing that machine learning algorithms can address similar problems that have been traditionally tackled in psychology using inferential statistics, my work initially sought to compare these frameworks and is now focused on creating new tools by focusing on the integration of machine learning and psychometrics.

  • Suzuki, H. L.* & Gonzalez, O. (2022). Relative predictive performance of treatments of ordinal outcome variables across machine learning algorithms and class distributions. Journal of Behavioral Data Science, 2, 1-26. DOI: 10.35566/jbds/v2n2/suzuki.
  • Gonzalez, O. (2021). Psychometric and machine learning approaches to reduce the length of scales. Multivariate Behavioral Research. DOI: 10.1080/00273171.2020.1781585.
  • Gonzalez, O. (2021). Psychometric and machine learning approaches for diagnostic assessment and tests of individual classification. Psychological Methods, 26, 236-254. DOI: 10.1037/met0000317.
  • Gonzalez, O., O’Rourke, H. P., Wurpts, I. C., & Grimm, K. J. (2018). Analyzing Monte Carlo simulation studies with Classification and Regression Trees. Structural Equation Modeling: A Multidisciplinary Journal, 25, 403-413. DOI: 10.1080/10705511.2017.1369353.

Role of measurement on statistical mediation conclusions

This research specialty seeks to address measurement issues that arise in mediation analysis, which is a key gap in the literature. Statistical mediation analysis is a statistical approach that identifies intermediate variables, known as mediators, that explain the causal relation between a predictor and an outcome. My focus is on the mediation by design paradigm, in which the relation between mechanisms and health outcomes is first identified (i.e., the conceptual theory of mediation), and then interventions are built to target those mechanisms (i.e., the action theory of mediation), which in turn should cause a change in the outcome. I argue that in order to identify mechanisms in a causal process and advance the theoretical understanding of phenomena, it is crucial to have clear interpretation and precise measurement of the (hypothesized) mechanisms involved.

  • Webb, C. A., Hirshberg, M. J., Gonzalez, O., Davidson, R. J., & Goldberg, S. B. (accepted). Revealing subgroup-specific mechanisms of change via moderated mediation: A meditation intervention example. Journal of Consulting and Clinical Psychology.
  • Valente, M. J., Rijnhart, J. J. M., & Gonzalez, O. (in press). A novel approach to estimating moderated treatment effects and moderated mediation effects with continuous moderators. Psychological Methods. DOI: 10.1037/met0000593
  • Gonzalez, O., Millechek, J. R.,* & Georgeson, A. R. (in press). Estimating latent baseline-by-treatment interactions in statistical mediation analysis. Structural Equation Modeling: A Multidisciplinary Journal. DOI: 10.1080/10705511.2023.2193312.
  • Gonzalez, O. & Valente, M. J. (in press). Accommodating a latent XM interaction in statistical mediation analysis. Multivariate Behavioral Research. DOI: 10.1080/00273171.2022.2119928.
  • Georgeson, A. R., Valente, M. J., & Gonzalez, O. (2022). The effect of partial invariance on the estimation of the mediated effect in the two-wave mediation model. Structural Equation Modeling: A Multidisciplinary Journal, 29, 909-918. DOI: 10.1080/10705511.2022.2067164.
  • Valente, M. J., Georgeson, A. R., & Gonzalez, O. (2021). Clarifying the implicit assumptions of two-wave mediation models via the latent change score specification. Frontiers in Psychology, 12, e709198. DOI: 10.3389/fpsyg.2021.709198.
  • Georgeson, A. R., Valente, M. J., & Gonzalez, O. (2021). Evaluating response shift in statistical mediation analysis. Advances in Methods and Practices in Psychological Science, 4, 1-13.  DOI: 10.1177/25152459211012271.
  • Gonzalez, O. & MacKinnon, D. P. (2021). The measurement of the mediator and its influence on statistical mediation conclusions. Psychological Methods, 26, 1-17. DOI: 10.1037/met0000263.
  • MacKinnon, D. P., Valente, M. J., & Gonzalez, O. (2020). The correspondence between causal and traditional mediation analysis: The link is in the mediator by treatment interaction. Prevention Science, 21, 147-157. DOI: 10.1007/s11121-019-0107604.
  • Olivera-Aguilar, M., Rikoon, S., Gonzalez, O., Kisbu-Sakarya, Y., & MacKinnon, D. P. (2018). Bias, Type 1 error rates, and statistical power of a latent mediation model in the presence of violations of invariance. Educational and Psychological Measurement, 78, 460-481. DOI: 10.1177/0013164416684169.
  • Gonzalez, O. & MacKinnon, D. P. (2018). A bifactor approach to model multifaceted constructs in statistical mediation analysis. Educational and Psychological Measurement, 78, 5-31. DOI:10.1177/0013164416673689.
  • Miočević, M., Gonzalez, O., Valente, M. J., & MacKinnon, D. P. (2018). A tutorial on Bayesian potential outcomes mediation analysis. Structural Equation Modeling: A Multidisciplinary Journal, 25, 121-136. DOI: 10.1080/10705511.2017.1342541.
  • Valente, M. J., Gonzalez, O., Miočević, M., & MacKinnon, D. P. (2016). A note on testing mediated effects in structural equation models: Reconciling past and current research on the performance of the test of joint significance. Educational and Psychological Measurement, 76, 889-911. DOI: 10.1177/0013164415618992.

Validation of assessments used for measurement and screening

Finally, I am also interested on the validation of psychological constructs. Validity refers to the amount of evidence and the theoretical rational that supports the interpretation and use of a score. Specifically, I am interested in the assessment of classification accuracy and consistency of screening measures, and how those statistics are affected by differential item functioning. Also, I am interested in studying construct overlap and jingle and jangle fallacies. The jangle fallacy indicates that two measures with different labels could actually assess the same construct, and the jingle fallacy indicates that two measures with the same label could assess the different construct. Finally, I am also interested in the evaluation of measures in the ontology of self-regulation (e.g., grit, mindfulness, etc).

  • Gonzalez, O., Georgeson, A. R., & Pelham, W. E. III. (2023). How accurate and consistent are score-based assessment decisions? A procedure using the linear factor model. Assessment, 30, 1640-1650. DOI: 10.1177/10731911221113568.
  • Gonzalez, O. (2023). Summary intervals for model-based classification accuracy and consistency indices. Educational and Psychological Measurement 83, 240-261. DOI: 10.1177/00131644221092347.
  • Gonzalez, O., Georgeson, A. R., Pelham, W. E. III., & Fouladi, R. T. (2021). Estimating classification consistency of screening measures and quantifying the impact of measurement bias. Psychological Assessment, 37, 596-609. DOI: 10.1037/pas0000938.
  • Gonzalez, O., MacKinnon, D. P., & Muniz, F. B., (2021). Extrinsic convergent validity evidence to prevent Jingle and Jangle fallacies. Multivariate Behavioral Research, 56, 3-19. DOI: 10.1080/00273171.2019.1707061.
  • Mazza, G. L., Smyth, H. L., Bissett, P. G., Canning, J. R., Eisenberg, I. W., Enkavi, A. Z., Gonzalez, O., Kim, J. S., Metcalf, S. A., Muniz, F., Onken, L, Pelham, W. E. III, Scherer, E. A., Stoeckel, L. E., Valente, M. J., Xie, H., Poldrack, R. A., Marsch, L. A., MacKinnon, D. P. (2021). Correlation database of 60 cross-disciplinary survey and cognitive tasks assessing self-regulation. Journal of Personality Assessment, 103, 238-245. DOI: 10.1080/00223891.2020.1732994.
  • Gonzalez, O. & Pelham, W. E. III. (2021). When does differential item functioning matter for screening? A method for empirical evaluation. Assessment, 28, 446-456. DOI: 10.1177/1073191120913618.
  • Gonzalez, O., Canning, J. R., Smyth, H., & MacKinnon, D. P. (2020). A psychometric evaluation of the Short Grit Scale: A closer look at its factor-structure and scale functioning. European Journal of Psychological Assessment, 36, 646-657. DOI: 10.1027/1015-5759/a000535.
  • Pelham, W. E. III, Gonzalez, O., Metcalf, S. A., Whicker, C. L., Witkiewitz, K. A., Marsch, L. A., & MacKinnon, D. P. (2019). Evaluating the factor structure of each facet of the Five Facets Mindfulness Questionnaire. Mindfulness, 10, 2629-2646. DOI: 10.1007/s12671-019-01235-2.
  • Pelham, W. E. III, Gonzalez, O., Metcalf, S. A., Whicker, C. L., Scherer, E. A., Witkiewitz, K., Marsch, L. A., Mackinnon, D. P. (2019). Item response theory analysis of the Five Facet Mindfulness Questionnaire and its short forms. Mindfulness10, 1615–1628. DOI: 10.1007/s12671-019-01105-x.
  • Eisenberg, I. W., Bissett, P. G., Canning, J. R., Dallery, J., Enkavi, A. Z., Gabrieli, S. W., Gonzalez, O., Green, A. I., Greene, M. A., Kiernan, M., Kim, S. J., Li, J., Lowe, M., Mazza, G. L., Metcalf, S. A., Onken, L., Peters, E., Prochaska, J. J., Scherer, E. A., Stoeckel, L. E., Valente, M. J., Xie, H., MacKinnon, D. P., Marsch, L. A., & Poldrack, R. A. (2018). Applying novel technologies and methods to inform the ontology of self-regulation. Behavior Research and Therapy. 101, 46-57. DOI: 10.1016/j.brat.2017.09.014.