Psychometric Data Analysis: Dimension Reduction of MHT Items Based on LASSO Regression

DOI

10.2991/978-94-6463-992-6_19How to use a DOI?

Keywords

Mental Health Test; item dimensionality reduction; LASSO regression; dataset expansion

Abstract

This study selected 1,109 middle school students as experimental subjects and conducted a survey using MHT (Mental Health Test), which includes eight psychological dimensions and one validity dimension, with a total of 100 items. The “Doctor Review Grade” from clinical diagnoses was used as the “gold standard” for criterion validity to construct a data analysis model based on students’ response data. Aiming at the issues of response fatigue and insufficient data validity caused by the large number of items in the traditional MHT, LASSO regression was employed for item dimensionality reduction. After data preprocessing, 904 valid samples were screened out. The accuracy of the original 90-item MHT model was 64.1%, with an AUC value of 0.697. After LASSO screening, 32 items were retained, and the model accuracy increased to 69.9%, with the AUC value rising to 0.737, achieving the goal of improving model accuracy while reducing the number of test items. Considering that the original test item scores were set as discrete values, to make students’ responses more accurately reflect their psychological states, we further expanded the dataset using a uniform distribution to convert item scores into continuous numerical variables. The results showed stable item selection, with the best model performance achieved when the number of samples reached approximately 10,000, the accuracy reached 70.8%, and the model evaluation indicators tending to stabilize when the sample number exceeded 30,000. At the end of this paper, the methodology and results employed in our study were discussed and summarized.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yirun Zhao
AU  - Junqing Yuan
AU  - Miaoshu Wu
PY  - 2026
DA  - 2026/02/20
TI  - Psychometric Data Analysis: Dimension Reduction of MHT Items Based on LASSO Regression
BT  - Proceedings of the 2025 4th International Conference on Mathematical Statistics and Economic Analysis (MSEA 2025)
PB  - Atlantis Press
SP  - 186
EP  - 194
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6463-992-6_19
DO  - 10.2991/978-94-6463-992-6_19
ID  - Zhao2026
ER  -