H1 predicts a performance interaction between task complexity and the formalization built into
the audit processes of the participants’ employing firm. To test this hypothesis, a multivariate
analysis of variance (MANOVA) model was estimated, with Performance on each of the four
experimental tasks as dependent variables, and Formalization, Task, and Complexity (nested within
Task) as independent variables. Task (1–4) and Complexity (high/low) are within-participant
variables, while Formalization (high/low) is a between-participants variable. Three betweenparticipant
covariates—Experience, Motivation, and Problem-Solving Ability (all continuous)—
were also considered in the analysis. Auditor Performance on each task is measured as the number
correct out of five possible points, as indicated previously. Results of the analysis are reported in
Table 2.