Scholarlink

<p>Despite the mounting demand for generative population models, their limited generalizability to underrepresented demographic groups hinders widespread adoption in real-world applications. Here we propose a diversity-aware population modeling framework that can guide targeted strategies in public health and education, by estimating subgroup-level effects and stratifying predictions to capture sociodemographic variability. We leverage Bayesian multilevel regression and post-stratification to systematically quantify inter-individual differences in the relationship between socioeconomic status and cognitive development. Post-stratification enhanced the interpretability of model predictions across underrepresented groups by incorporating US Census data to gain additional insights into smaller subgroups in the Adolescent Brain Cognitive Development Study. This ensured that predictions were not skewed by overly heterogeneous or homogeneous representations. Our analyses underscore the importance of combining Bayesian multilevel modeling with post-stratification to validate reliability and provide a more holistic explanation of sociodemographic disparities in our diversity-aware population modeling framework.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Quantifying associations between socio-spatial factors and cognitive development in the ABCD cohort

  • 期刊论文

  • 发表日期：

    2025年3月20日

查看全文
  • Nicole Osayande,
  • Justin Marotta,
  • Shambhavi Aggarwal,
  • Jakub Kopal,
  • Avram Holmes,
  • Sarah W. Yip,
  • Danilo Bzdok

摘要

Despite the mounting demand for generative population models, their limited generalizability to underrepresented demographic groups hinders widespread adoption in real-world applications. Here we propose a diversity-aware population modeling framework that can guide targeted strategies in public health and education, by estimating subgroup-level effects and stratifying predictions to capture sociodemographic variability. We leverage Bayesian multilevel regression and post-stratification to systematically quantify inter-individual differences in the relationship between socioeconomic status and cognitive development. Post-stratification enhanced the interpretability of model predictions across underrepresented groups by incorporating US Census data to gain additional insights into smaller subgroups in the Adolescent Brain Cognitive Development Study. This ensured that predictions were not skewed by overly heterogeneous or homogeneous representations. Our analyses underscore the importance of combining Bayesian multilevel modeling with post-stratification to validate reliability and provide a more holistic explanation of sociodemographic disparities in our diversity-aware population modeling framework.