SiRJ

SiRJ models cognitive processes underlying responses to situational judgment tests (SJTs) to analyze how conditional reasoning, similarity judgments, and preference accumulation produce observed response patterns.

Key Features:

• Process-oriented framework: Positions SJT responses as arising from a sequence of cognitive operations rather than solely from latent psychometric constructs.
• Specified cognitive operations: Explicitly models conditional reasoning, similarity judgments, and preference accumulation as components of respondent decision-making.
• Formal computational model: Translates the SiRJ framework into a formal computational model with accompanying simulation code.
• Simulation study: Uses simulation studies to validate the model and reproduce empirical patterns in SJT data.
• Neural network algorithms: Employs neural network algorithms for parts of the simulation and analysis.
• Bayesian survival analysis: Utilizes Bayesian survival analysis as an analytical approach within the simulation study.
• Empirical replication and insight generation: Replicates existing empirical effects in SJTs and generates new hypotheses for interpretation and item development.
• Item and instruction effects: Addresses how variations in item construction and instructions influence respondent choices.
• Psychological and contextual factors: Accounts for psychological and contextual factors that modulate judgment and decision-making in SJT scenarios.

Scientific Applications:

• Interpreting SJT responses: Decomposes SJT responses into component cognitive operations to inform interpretation beyond latent-construct psychometric models.
• Informing item development and scoring: Guides development and scoring of SJT items by revealing process-level determinants of choices.
• Evaluating item and instruction effects: Evaluates how changes in item construction and instructions alter response patterns.
• Hypothesis generation and validation: Generates and tests hypotheses via simulation to replicate empirical effects and predict new outcomes.
• Studying judgment and decision-making: Supports research on context-specific judgment and decision-making processes implicated in SJTs.

Methodology:

The framework is translated into a formal computational model and evaluated via simulation studies using neural network algorithms and Bayesian survival analysis.