Xingyao (Doria) Xiao

All Publications

• Revisiting reliability with human and machine learning raters under scoring design and rater configuration in the many-facet Rasch model. The British journal of mathematical and statistical psychology Xiao, X., Patz, R. J., Wilson, M. R. 2026

  Abstract

  Constructed-response (CR) items are widely used to assess higher order skills but require human scoring, which introduces variability and is costly at scale. Machine learning (ML)-based scoring offers a scalable alternative, yet its psychometric consequences in rater-mediated models remain underexplored. This study examines how scoring design, rater bias, ML inconsistency and model specification affect the reliability of ability estimation in polytomous CR assessments. Using Monte Carlo simulation, we manipulated human and ML rater bias, ML inconsistency and scoring density (complete, overlapping, isolated). Five estimation models were compared, including the Partial Credit Model (PCM) with fixed thresholds and the Many-Facet Partial Credit Model (MFPCM) with and without free calibration. Results showed that systematic bias, not random inconsistency, was the main source of error. Hybrid human-ML scoring improved estimation when raters were unbiased or exhibited opposing biases, but error compounded when biases aligned. Across designs, PCM with fixed thresholds consistently outperformed more complex alternatives, while anchoring CR items to selected-response metrics stabilized MFPCM estimation. The real data application replicated these patterns. Findings show that scoring design and bias structure, rather than model complexity, drive the benefits of hybrid scoring and that anchoring offers a practical strategy for stabilizing estimation.

  View details for DOI 10.1111/bmsp.70034

  View details for PubMedID 41618685

• Gendered pathways to self-efficacy: moderating and mediating roles of family, school, and sibling contexts in early adolescence EUROPEAN JOURNAL OF PSYCHOLOGY OF EDUCATION Xiao, X., Cheng, Y. 2025; 41 (1)

  View details for DOI 10.1007/s10212-025-01037-2

  View details for Web of Science ID 001642766000001

• Trajectories of Depressive Symptom Among College Students in China During the COVID-19 Pandemic: Association With Suicidal Ideation and Insomnia Symptoms. Suicide & life-threatening behavior Su, B., Xiao, X., Cheng, Y., Liu, C., Yang, C. 2025; 55 (5): e70051

  Abstract

  Despite the burgeoning literature on mental health problems during the COVID-19 pandemic, little is understood about the heterogeneous developmental trajectories of depressive symptoms and their subsequent mental health outcomes.This three-year, five-wave longitudinal study aimed to identify distinct trajectories of depressive symptoms among Chinese college students during the COVID-19 pandemic. We further examined the impact of these different trajectories on suicidal ideation and insomnia symptoms.Participants included 1387 Chinese college students (27% male, average age = 19.24, SD = 0.99) from a five-wave longitudinal online survey. Data were analyzed using Growth Mixture Modeling (GMM), a person-centered approach, to identify distinct symptom trajectories.Five distinct trajectories of depression were identified: resilient (24.01%), moderate-remission (46.27%), low-increasing (19.68%), high-recover (5.55%), and moderate-increasing (4.35%). The analysis demonstrated that these trajectories of depression symptoms effectively predicted changes in suicidal ideation and insomnia symptoms over time.These findings highlight the significant heterogeneity of depression trajectories among Chinese college students and their strong association with critical mental health outcomes. The results suggest that individuals following increasing-symptom trajectories are at particular risk for negative outcomes like suicidality and sleep disturbances.Therefore, mental health service workers should not only focus on individuals' temporary state of depression but also monitor shifts in their depressive symptoms over time. Early identification of adverse trajectories can inform targeted interventions to mitigate the risk of severe outcomes like suicidal ideation and insomnia.

  View details for DOI 10.1111/sltb.70051

  View details for PubMedID 40947922

• Automatic Prompt Engineering for Automatic Scoring JOURNAL OF EDUCATIONAL MEASUREMENT Xue, M., Liu, Y., Xiao, X., Wilson, M. 2025

  View details for DOI 10.1111/jedm.70002

  View details for Web of Science ID 001551213800001

• Bayesian Identification and Estimation of Growth Mixture Models PSYCHOMETRIKA Xiao, X., Rabe-Hesketh, S., Skrondal, A. 2025; 90 (2): 442-475

  View details for DOI 10.1017/psy.2025.11

  View details for Web of Science ID 001514063000001

  View details for PubMedID 40190064

• Patterns of participation and performance at the class level in English online education: A longitudinal cluster analysis of online K-12 after-school education in China EDUCATION AND INFORMATION TECHNOLOGIES Wang, F., Zhu, X., Pi, L., Xiao, X., Zhang, J. 2024; 29 (12): 15595-15619

  View details for DOI 10.1007/s10639-024-12451-2

  View details for Web of Science ID 001152699900002

• A multi-level analysis on the causes of train-pedestrian collisions in Southwest China 2011-2020. Accident; analysis and prevention Zhang, J., Liu, F., Chen, Z., Yu, Z., Xiao, X., Shi, L., Guo, Z. 2023; 193: 107332

  Abstract

  Collisions between trains and pedestrians are the primary cause of railway casualties. However, there remains a lack of comprehensive understanding regarding the underlying causes of this phenomenon. This study employs a multi-level approach to investigate the factors associated with the occurrence and severity of train-pedestrian collisions. The investigation is based on 2160 independent cases that occurred in southwest China from 2011 to 2020. Multiple contributing factors related to the victim, train, track, and socio-economic status of the surrounding district were examined, utilizing information from various sources. At the county level, several risk factors were identified in predicting the occurrence rate. These factors include higher population density and a greater number of normal-speed stations. However, the presence of high-speed train stations did not exhibit any significant impact. Additionally, the study found that regulations pertaining to protective fences were highly effective in reducing the occurrence rate. Regarding the prediction of collision severity, certain factors were found to increase the death rate. These factors include young men as victims, engaging in lying down or crossing behaviors, higher train speeds, gentle downhill slopes, lower education levels, and a higher proportion of the labor force. These findings emphasize the necessity of adopting a comprehensive perspective when examining the causes of train-pedestrian collisions. Furthermore, it underscores the significance of considering the notable differences between rapidly developing countries such as China and developed countries. Based on our findings, we also provide corresponding policy suggestions.

  View details for DOI 10.1016/j.aap.2023.107332

  View details for PubMedID 37801815

• Movie Title Keywords: A Text Mining and Exploratory Factor Analysis of Popular Movies in the United States and China JOURNAL OF RISK AND FINANCIAL MANAGEMENT Xiao, X., Cheng, Y., Kim, J. 2021; 14 (2)

  View details for DOI 10.3390/jrfm14020068

  View details for Web of Science ID 000622686900001

• Directional dependence between major cities in China based on copula regression on air pollution measurements. PloS one Kim, J. M., Lee, N., Xiao, X. 2019; 14 (3): e0213148

  Abstract

  Air pollution is well-known as a major risk to public health, causing various diseases including pulmonary and cardiovascular diseases. As social concern increases, the amount of air pollution data is increasing rapidly. The purpose of this study is to statistically characterize dependence between major cities in China based on a measure of directional dependence estimated from PM2.5 measurements. As a measure of the directional dependence, we propose the so-called copula directional dependence (CDD) using beta regression models. An advantage of the CDD is that it does not rely on strict assumptions of specific probability distributions or linearity. We used hourly PM2.5 measurement data collected at four major cities in China: Beijing, Chengdu, Guangzhou, and Shanghai, from 2013 to 2017. After accounting for autocorrelation in the PM2.5 time series via nonlinear autoregressive models, CDDs between the four cities were estimated to produce directed network structures of statistical dependence. In addition, a statistical method was proposed to test the directionality of dependence between each pair of cities. From the PM2.5 data, we could discover that Chengdu and Guangzhou are the most closely related cities and that the directionality between them has changed once during 2013 to 2017, which implies a major economic or environmental change in these Chinese regions.

  View details for DOI 10.1371/journal.pone.0213148

  View details for PubMedID 30870434

  View details for PubMedCentralID PMC6417661