Order-constrained inference: a nuanced approach to hypothesis testing

Many statistical analyses performed in psychological studies add extraneous assumptions that are not part of the theory. These added assumptions could adversely influence the conclusions one derives from the analyses. Order-constrained inference allows researchers to avoid unnecessary assumptions, translate verbal predictions into direct testable hypotheses, and run model selection among competing theories. We reanalyzed data from two separate case studies to highlight how one can use order-constrained modeling to formulate more nuanced hypotheses and test these hypotheses jointly. To further leverage order-constrained inference, we compared the performance of competing theories using Bayesian model selection methods in the second case study. We observe that order-constrained inference not only provides us with a coarse view of all the hypotheses at the joint level, it also offers a fine-grained perspective of all the hypotheses at the item level that might otherwise stay hidden if we only assessed trends at the aggregate level.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Psychology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2024

Author dc:creator

• Chen, Meichai

Contributors dc:contributor

• Regenwetter, Michel
• Koehn, Hans Friedrich

• Order-constrained Inference
• Nuanced Hypotheses
• Model Competition

Statement dc:rights

• Copyright 2024 Meichai Chen

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/124465

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F124465