{"response":{"docs":[{"system_create_dtsi":"2026-06-02T23:02:05Z","system_modified_dtsi":"2026-06-03T14:52:45Z","has_model_ssim":["Article"],"id":"x920fz71r","accessControl_ssim":["2f435893-9a60-4fbf-b57e-7f0144dbc692"],"hasRelatedMediaFragment_ssim":["q524jq547"],"hasRelatedImage_ssim":["q524jq547"],"depositor_ssim":["kelceybn@ucmail.uc.edu"],"depositor_tesim":["kelceybn@ucmail.uc.edu"],"title_tesim":["Targeted After Generative (TAG) Learning for Causal Inference with Latent Variables"],"date_uploaded_dtsi":"2026-06-02T23:02:01Z","date_modified_dtsi":"2026-06-03T14:52:45Z","isPartOf_ssim":["admin_set/default"],"doi_tesim":["doi:10.7945/gr8r-j398"],"college_tesim":["Education, Criminal Justice, and Human Services"],"department_tesim":["Quantitative Research Methodology"],"creator_tesim":["Kelcey, Ben","Ataneka, Amota"," Habarurema, Jean-Baptiste"],"publisher_tesim":["University of Cincinnati"],"subject_tesim":["Structural equation modeling, deep generative learning, targeted learning, variational autoencoders"],"description_tesim":["Latent variables and causal inference are central to implementing empirical research and advancing substantive theory across the social sciences. Conventional latent variable frameworks such as structural equation modeling (SEM) are constrained by several simplifying assumptions (e.g., linearity, additivity, correct model specification). In this study, we relax key SEM assumptions by building data adaptive estimation methods for causal inference with latent variables. We developed a theory-driven Targeted After Generative (TAG) learning framework that integrates SEM-informed variational autoencoders (SEM VAEs) with targeted machine learning to estimate the average treatment effect. On the generative side, the SEM VAE encodes theory aligned neural networks that learn latent variable distributions through flexible joint modeling of measurement and structural parameters. On the targeted learning side, we construct an efficient influence function for the treatment effect with the SEM-VAE implied posterior distribution of the latent variables to estimate the average effect. Across simulations, the TAG learning approach outperforms alternative approaches and demonstrates low to no bias even in adverse conditions. TAG learning offers a theory-driven flexible approach with methodological promise for causally interpretable data-adaptive analysis with latent variables. More practically, the framework allows estimation of causal effects involving latent variables even in the presence of unknown functional relationships in the measurement and structural models (e.g., data adaptive effect estimation for nonlinear/nonadditive SEMs)."],"license_tesim":["http://creativecommons.org/licenses/by/4.0/"],"date_created_tesim":["05-20-26"],"thumbnail_path_ss":"/downloads/q524jq547?file=thumbnail","suppressed_bsi":false,"actionable_workflow_roles_ssim":["admin_set/default-default-depositing"],"workflow_state_name_ssim":["deposited"],"member_ids_ssim":["q524jq547","fn1070797"],"file_set_ids_ssim":["q524jq547","fn1070797"],"visibility_ssi":"open","admin_set_tesim":["Default Admin Set"],"sort_title_ssi":"TARGETED AFTER GENERATIVE TAG LEARNING FOR CAUSAL INFERENCE WITH LATENT VARIABLES","human_readable_type_tesim":["Article"],"read_access_group_ssim":["public"],"edit_access_group_ssim":["admin"],"edit_access_person_ssim":["kelceybn@ucmail.uc.edu"],"nesting_collection__pathnames_ssim":["x920fz71r"],"nesting_collection__deepest_nested_depth_isi":1,"_version_":1866987860078362624,"timestamp":"2026-06-03T14:52:51.199Z","score":0.00049999997}],"facets":[{"name":"human_readable_type_sim","items":[{"value":"Article","hits":1,"label":"Article"}],"label":"Human Readable Type Sim"},{"name":"creator_sim","items":[{"value":" Habarurema, Jean-Baptiste","hits":1,"label":" Habarurema, Jean-Baptiste"},{"value":"Ataneka, Amota","hits":1,"label":"Ataneka, Amota"},{"value":"Kelcey, Ben","hits":1,"label":"Kelcey, Ben"}],"label":"Creator Sim"},{"name":"subject_sim","items":[{"value":"Structural equation modeling, deep generative learning, targeted learning, variational autoencoders","hits":1,"label":"Structural equation modeling, deep generative learning, targeted learning, variational autoencoders"}],"label":"Subject Sim"},{"name":"college_sim","items":[{"value":"Education, Criminal Justice, and Human Services","hits":1,"label":"Education, Criminal Justice, and Human Services"}],"label":"College Sim"},{"name":"department_sim","items":[{"value":"Quantitative Research Methodology","hits":1,"label":"Quantitative Research Methodology"}],"label":"Department Sim"},{"name":"language_sim","items":[],"label":"Language Sim"},{"name":"publisher_sim","items":[{"value":"University of Cincinnati","hits":1,"label":"University of Cincinnati"}],"label":"Publisher Sim"},{"name":"date_created_sim","items":[{"value":"05-20-26","hits":1,"label":"05-20-26"}],"label":"Date Created Sim"},{"name":"member_of_collection_ids_ssim","items":[],"label":"Member Of Collection Ids Ssim"},{"name":"generic_type_sim","items":[{"value":"Work","hits":1,"label":"Work"}],"label":"Generic Type Sim"}],"pages":{"current_page":1,"next_page":null,"prev_page":null,"total_pages":1,"limit_value":10,"offset_value":0,"total_count":1,"first_page?":true,"last_page?":true}}}