The article “Rejoinder to McNeish and Mislevy: What Does Psychological Measurement Require?” by Klaas Sijtsma, Jules L. Ellis, and Denny Borsboom provides a detailed response to criticisms and discussions raised by McNeish and Mislevy regarding the role and application of the sum score in psychometric practices. The authors address core concerns while emphasizing the need for a balance between advanced psychometric techniques and practical, transparent approaches.
Background
The rejoinder builds on discussions about the use of the sum score versus more sophisticated latent variable models like item response theory (IRT) in psychological measurement. The authors argue for the responsible use of both methods, stressing that the sum score remains a useful tool in contexts where simplicity and transparency are essential. The conversation highlights a growing divide between psychology and psychometrics, urging collaboration to strengthen theoretical and practical foundations in the field.
Key Insights
- The Role of Psychometrics Education and Outreach: The authors emphasize the need for enhanced education and outreach to help researchers responsibly use advanced methods like IRT. They highlight that while training is essential, direct collaboration between researchers and psychometricians is crucial for practical application.
- Sum Score Versus Latent Variable Models: The article defends the sum score as a practical and transparent scoring method, particularly in scenarios where granular precision is unnecessary. The authors discuss its limitations but argue for its continued relevance alongside latent variable models.
- Bridging Psychology and Psychometrics: A key theme is the growing gap between psychology and psychometrics. The authors call for better integration of psychological theory with psychometric practices to enhance the validity and utility of measurement tools.
Significance
This work underscores the importance of aligning psychometric advancements with practical needs in psychological testing. By advocating for a dual approach that incorporates both sum scores and latent variable models, the authors address concerns about oversimplification without dismissing the value of transparency. The discussion contributes to ongoing debates about measurement theory, education, and application in psychology.
Future Directions
The authors suggest several areas for further research, including improving collaboration between psychology and psychometrics, exploring methods to balance simplicity and precision in scoring, and investigating the implications of machine learning and data-driven approaches in measurement and prediction. They also highlight the need for more robust theoretical development in psychological attributes.
Conclusion
The article effectively addresses concerns raised by McNeish and Mislevy, advocating for a balanced perspective on psychometric methods. By emphasizing education, outreach, and collaboration, the authors make a compelling case for bridging the divide between psychology and psychometrics to enhance the field’s theoretical and practical contributions.
Reference
Sijtsma, K., Ellis, J. L., & Borsboom, D. (2024). Rejoinder to McNeish and Mislevy: What Does Psychological Measurement Require? Psychometrika, 89, 11175–1185 (2024). https://doi.org/10.1007/s11336-024-10004-7
