QTest: Quantitative testing of theories of binary choice.

Abstract

The goal of this paper is to make modeling and quantitative testing accessible to behavioral decision researchers interested in substantive questions. We provide a novel, rigorous, yet very general, quantitative diagnostic framework for testing theories of binary choice. This permits the nontechnical scholar to proceed far beyond traditionally rather superficial methods of analysis, and it permits the quantitatively savvy scholar to triage theoretical proposals before investing effort into complex and specialized quantitative analyses. Our theoretical framework links static algebraic decision theory with observed variability in behavioral binary choice data. The paper is supplemented with a custom-designed public-domain statistical analysis package, the QTest software. We illustrate our approach with a quantitative analysis using published laboratory data, including tests of novel versions of “Random Cumulative Prospect Theory.” A major asset of the approach is the potential to distinguish decision makers who have a fixed preference and commit errors in observed choices from decision makers who waver in their preferences.

Mate choice in fruit flies is rational and adaptive

“…Although assessing transitivity via triads is powerful and intuitive, the simplification of the network into a number of relationship subsets can miss patterns operating on the network as a whole213536. We therefore validated our results via analysis of longer chains of relationships following Regenwetter et al 21.…”

Individual differences in the algebraic structure of preferences

“…Harless & Camerer, 1994). Other possible specifications include ''distance-based'' and combinations of mixture and error, -see Regenwetter et al (2014) for a more complete discussion of alternative stochastic specifications.…”

Bayes factors for state-trace analysis

“…In other words, P can be described as a weighted sum such that each a j weight corresponds to the probability of selecting an individual who chooses according to the corresponding ordering relationship. See also Regenwetter et al (2014) for a discussion of how convex hull models can be interpreted using averaged data. Conversely, if the aggregated group-level data, P, are not well described by this convex hull model then these data cannot be described via a probability distribution over individuals who all produce responses that satisfy one of the viable orders.…”