Haonan Yin

Haonan Yin

Auto-generated profile, based on publicly available information

Haonan Yin

Department of Information Systems and Business Analytics, Ivy College of Business, Iowa State University , Ames, IA 50011

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 March 2025 in Journal of Medical Internet Research
Verified
Journal Article
Open access
HKHuawei Kuang
Huawei Kuang
LZLichao Zhu
Lichao Zhu
HYHaonan Yin
Haonan Yin
ZZZihe Zhang
Zihe Zhang
BJBiao Jing
Biao Jing
JKJunwei Kuang
Junwei Kuang
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Online questionnaires are widely used for large-scale mental health screening. However, careless responses from participants can compromise data validity, ultimately undermining the reliability of screening outcomes. Prior studies have primarily focused on the effects of individual and environmental factors on careless responding, the heterogenous effect role of questionnaire type remains underexplored. Furthermore, few studies have examined careless responding within mental health screening contexts. Objective: This study aims to investigate the individual factors influencing careless responses in online mental health screening and assess how the effect of these factors varies across different types of psychological screening instruments. Methods: This study analyzes data from 24,367 participants at Xinxiang Medical University across four psychological questionnaires (PHQ-9, PSS, ISI, GAD-7). Careless responding is defined as a continuous dependent variable, measured by the proportion of items completed in less than two seconds per item — a widely accepted criterion for identifying insufficient response time. We employ multiple linear regression model to examine the effect of individual factors (gender, age, education, smoking, and drinking) on careless responding while assessing variations across questionnaire types. Additionally, response time distributions are visualized to identify typical patterns between careless and careful responders across four psychological questionnaires. Results: Females consistently demonstrate lower levels of careless responding than males, when completing the PHQ-9 (β=-0.172, 95% CI -0.104 to -0.089, P<.001), PSS (β=-0.234, 95% CI -0.162 to -0.144, P<.001), ISI (β=-0.207, 95% CI -0.13 to -0.114, P<.001), and GAD-7 (β=-0.177, 95% CI -0.108 to -0.093, P<.001). While older participants demonstrate lower levels of careless responding when completing the PHQ-9 (β=-0.036, 95% CI -0.007 to -0.003], P<.001), ISI (β=-0.036, 95% CI -0.007 to -0.003, P<.001) and GAD-7 (β=-0.053, 95% CI -0.009 to -0.005], P<.001), their age is unrelated to careless responding on the PSS. Interestingly, compared with participants with a associate-level education, high education background (bachelor, master, doctor) consistently demonstrates high-level of careless responding, especially for master’s degree holders (PHQ-9: β=0.098, 95% CI 0.136 to 0.188], P<.001; GAD-7: β=0.091, 95% CI 0.125 to 0.178, P<.001). Smokers exhibit varied patterns, with current smokers demonstrate lower levels of careless responding the PHQ-9 (β=-0.022, 95% CI -0.064 to -0.016, P=.001) and GAD-7 (β=-0.014, 95% CI -0.051 to -0.002, P=.03) but occasional smokers demonstrate higher levels of careless responding the PSS (β=0.019, 95% CI 0.010 to 0.050, P=.003) than non-smokers. Drinkers consistently demonstrate lower levels of careless responding than non-drinkers, with the strongest effect for drinks occasionally on the PHQ-9 (β=-0.163, 95% CI -0.103 to -0.087, P<.001). Analysis of response times revealed that participants tended to spend less time on surveys on depression and anxiety, and careless responding in surveys on stress and insomnia was characterized by skipping questions. Conclusions: The effect of individual factors varies across questionnaire types. These findings offer valuable insights for questionnaire designers and administrators, highlighting the need for targeted interventions tailored to user group characteristics and the unique features of different questionnaire types to mitigate careless responding.

17 January 2021 in SSRN Electronic Journal
Verified
Preprint
Open access
PDF file
NHNi Huang
Ni Huang
ZYZhijun Yan
Zhijun Yan
HYHaonan Yin
Haonan Yin
E-healthcare platforms start to integrate medical services across online and offline channels, where providers can perform online consultations, schedule patients’ offline visits, synchronize relevant medical records, and finally, answer online inquiries regarding follow-up and recovery anytime and anywhere within one operations management function (i.e., online-offline service integration). In this study, we seek to quantify the effects of online-offline service integration on the e-healthcare providers’ demand and reputational outcomes, noting that it is not altogether clear how the service integration function will affect the providers who adopt such a function in e-healthcare platforms. Leveraging a quasi-natural experiment on an e-healthcare platform, we conducted difference-in-differences analyses in tandem with a variety of matching strategies, including propensity score matching and look-ahead propensity score matching. Further, we explored the moderating roles of provider-specific characteristics. Our results reveal a set of robust and interesting findings: i) e-healthcare providers, on average, experience increases in online demand and decreases in offline demand post online-offline service integration; ii) the service integration function also improves the professional reputation of participating providers; iii) the impact of channel integration on the outcomes are weaker for providers with lower (vs. higher) professional titles; and iv) the providers who specialize in treating chronic (vs. acute) diseases experience greater increases in online demand and reputational outcomes, yet insignificant changes in offline demand. This work contributes to related prior literatures on healthcare operations management, e-healthcare, and online-offline channel integration, offering design implications for the service operations of e-healthcare platforms.

3 August 2026 in PNAS Nexus
Verified
Journal Article
Open access
HYHaonan Yin
Haonan Yin
SVShai Vardi
Shai Vardi
VCVidyanand Choudhary
Vidyanand Choudhary
Large language models (LLMs) are increasingly deployed in decision-support systems for high-stakes domains such as hiring and university admissions, where choices often involve selecting among competing alternatives. While prior work has noted position biases in LLM-driven comparisons, these biases have not been systematically analyzed or linked to underlying preference structures. We present the first comprehensive study of position biases across multiple LLMs and two distinct domains: resume comparisons, representing a realistic high-stakes context, and color selection, which isolates position effects by removing confounding factors. We find strong and consistent order effects, including a quality-dependent shift: when all options are high quality, models favor the first option, but when quality is lower, they favor later options. We also identify a previously undocumented bias: a name bias, where certain names are favored despite controlling for demographic signals. To separate superficial tie-breaking from genuine distortions of judgment, we extend the rational choice framework to classify pairwise preferences as robust, fragile, or indifferent. Using this framework, we show that order effects can lead models to select strictly inferior options. These results indicate that LLMs exhibit distinct failure modes not documented in human decision-making. We also propose targeted mitigation strategies, including a novel use of the temperature parameter, to recover underlying preferences when order effects distort model behavior.