Review of choice-based, matched, and other stratified sample studies in auditing research

Abstract

The use of choice-based, matched, and other stratified sample designs is common in auditing research. However, it is not widely appreciated that the data analysis for these studies has to take into account the non-random nature of sample selection in these designs. A choice-based, matched or otherwise stratified sample is a nonrandom sample that must be analyzed using conditional analysis techniques. We review five research streams in the auditing area. These streams include work on determinants of audit litigation, audit fees, auditor reporting in financially distressed firms, audit quality and auditor switches. Cram, Karan, and Stuart (CKS) (2009) demonstrated the accuracy of conditional analysis, compared to unconditional analysis, of nonrandom samples through the use of simulations, replications, and mathematical proofs. Papers since published have continued to rely upon questionable research, however, and it is hard for researchers to identify what is the reliability of a given work. We complement and extend CKS (2009) by identifying audit papers in selected research streams whose results will likely differ if the data gathered are analyzed using conditional analysis techniques. Thus research can be advanced either by replication and reanalysis, or by refocus of new research upon issues that should no longer be viewed as settled.

The Effect of Shareholder-Level Taxes on Organizational Form and Stock Ownership: Evidence from Equity Carve-Outs of Master Limited Partnerships

“…This suggests that my main results likely do not suffer from the potential problems noted in Cram et al (2009) and Stuart et al (2013).…”

“…While conditional logits are rarely performed in accounting research, Cram, Karan, and Stuart (2009) and Stuart, Shin, Cram, and Karan (2013) argue that conditional logit is the proper method for analyzing matched samples of data. Using simulations, Cram et al (2009) show that failure to use conditional logit when dealing with matched data can lead to incorrect acceptance or rejection of the null hypothesis.…”

The Role of Timing and Management's Remediation Actions in Preventing Failed Remediation of Material Weaknesses in Internal Controls*

“…Measuring the timing, nature, and extent of actions requires hand-collection of remediation disclosures. Choice-based matched samples are frequently used when data collection is costly, especially when outcomes are rare (Cram et al 2009;Stuart et al 2013). The use of a matched sample design allows me to focus power on estimating parameters for the timing, extent, and nature of remediation strategy while controlling for other variables that are not the focus of my study.…”

“…Therefore, it is unclear what a credible first stage prediction model would include. Stuart et al (2013) and Cram et al (2009) conclude that methods such as propensity score matching are not necessary when using a choicebased fully matched sample for a binary dependent variable as long as conditional logistic regression is used. 15.…”

Oportunidades para la investigación contable en Colombia

“…Los datos carecen de independencia y no cumplen algunos de los requisitos para las regresiones lineales (Chang, Lin, Liu, Shiue, & Wheatley, 2018). Por esta razón, muchos trabajos que desarrollan este enfoque usan métodos sofisticados, como muestras estratificadas (Stuart, Shin, Cram, & Karan, 2013), análisis de metarregresión (Hay & Knechel, 2017) o modelo lineal jerárquico (Chang, Lin, Liu, Shiue, & Wheatley, 2018), en busca de una mayor credibilidad para las investigaciones llevadas a cabo en auditoría. Recientemente se publicó un libro que presenta las bases de datos con información disponible para llevar a cabo investigaciones en este campo (Weirich, Churyk, & Pearson, 2015).…”