The 'Big N' Audit Quality Kerfuffle

Published on Oct 13, 2021in Social Science Research Network
· DOI :10.2139/SSRN.3511585
William M. Cready18
Estimated H-index: 18
(UTD: University of Texas at Dallas)
In a highly influential analysis, Lawrence, Minutti-Meza, and Zhang (2011), LMZ henceforth, report that statistically significant relations between a firm’s choice of a Big N auditor and three audit quality metrics (discretionary accruals, cost equity capital, and analyst forecast accuracy) turn “insignificant” after application of matching (propensity score and size) designs. LMZ, however, in interpreting these outcomes mistakenly identify the difference between statistically significant and statistically insignificant as significant (Gelman and Stern, 2006). This analysis re-examines the LMZ evidence descriptively. It finds that little descriptive support exists in the LMZ evidence for conclusive assertions regarding the “insignificance” of audit quality proxy level differences between Big N and non-Big N auditors. Nor does its evidence provide a reliable basis for thinking that propensity score matching based assessment of these differences produces substantially closer to zero inferences about them relative to inferences obtained from existent (inclusive of LMZ provided estimates) conventional non-matching design based multiple regression assessments. Indeed, the LMZ evidence is most appropriately interpreted as providing broad robustness support for the insights provided by such models.
📖 Papers frequently viewed together
83 Citations
2 Citations
3 Authors (Dominik Burger, ..., Andreas Blumer)
#1William M. Cready (UTD: University of Texas at Dallas)H-Index: 18
A field report on the non-impact of the ASA Statement on Statistical Significance and P-Values on the broader research community. This article discusses the implications of a premier accounting journal’s dismissive response to a systematic application of the 2016 ASA Statement principles to its published content for the ASA statement’s likely success in achieving its objective of improving statistical practice.
5 CitationsSource
#1Valentin Amrhein (University of Basel)H-Index: 27
#2David Trafimow (NMSU: New Mexico State University)H-Index: 50
Last. Sander Greenland (UCLA: University of California, Los Angeles)H-Index: 130
view all 3 authors...
AbstractStatistical inference often fails to replicate. One reason is that many results may be selected for drawing inference because some threshold of a statistic like the P-value was crossed, leading to biased reported effect sizes. Nonetheless, considerable non-replication is to be expected even without selective reporting, and generalizations from single studies are rarely if ever warranted. Honestly reported results must vary from replication to replication because of varying assumption vio...
80 CitationsSource
#1Ronald L. Wasserstein (American Statistical Association)H-Index: 4
#2Allen L. Schirm (Mathematica Policy Research)H-Index: 2
Last. Nicole A. Lazar (UGA: University of Georgia)H-Index: 14
view all 3 authors...
714 CitationsSource
#1Valentin AmrheinH-Index: 27
#2Sander GreenlandH-Index: 130
Last. Blake McShaneH-Index: 1
view all 3 authors...
Valentin Amrhein, Sander Greenland, Blake McShane and more than 800 signatories call for an end to hyped claims and the dismissal of possibly crucial effects. Valentin Amrhein, Sander Greenland, Blake McShane and more than 800 signatories call for an end to hyped claims and the dismissal of possibly crucial effects.
820 CitationsSource
#1William M. Cready (UTD: University of Texas at Dallas)H-Index: 18
#2Jiapeng He (UTD: University of Texas at Dallas)H-Index: 1
Last. Yang Zhang (UTD: University of Texas at Dallas)H-Index: 1
view all 6 authors...
This study evaluates how accounting researchers analyze and report null outcomes based on a large sample examination of recent accounting research publications. As null outcomes reflect an inability to reject a null they, unlike rejections, do not lend themselves to specifically conclusive interpretations. Rather, drawing useful inference from them requires fundamental descriptive analysis. In the 35 articles we identify as presenting substantive null outcomes, however, inappropriately conclusiv...
9 CitationsSource
#1Mark L. DeFondH-Index: 44
#2David H. ErkensH-Index: 11
Last. Jieying ZhangH-Index: 13
view all 3 authors...
A large auditing literature concludes that Big N auditors provide higher audit quality than non-Big N auditors. Recently, however, a high-profile study suggests that propensity score matching (PSM) on client characteristics eliminates the Big N effect [Lawrence A, Minutti-Meza M, Zhang P (2011) Can Big 4 versus non-Big 4 differences in audit-quality proxies be attributed to client characteristics? Accounting Rev. 86(1):259–286]. We conjecture that this finding may be affected by PSM’s sensitivit...
83 CitationsSource
#1Ronald L. Wasserstein (American Statistical Association)H-Index: 4
#2Nicole A. Lazar (American Statistical Association)H-Index: 2
2,313 CitationsSource
#1Sander Greenland (UCLA: University of California, Los Angeles)H-Index: 130
#2Stephen SennH-Index: 45
Last. Douglas G. Altman (University of Oxford)H-Index: 268
view all 7 authors...
Misinterpretation and abuse of statistical tests, confidence intervals, and statistical power have been decried for decades, yet remain rampant. A key problem is that there are no interpretations of these concepts that are at once simple, intuitive, correct, and foolproof. Instead, correct use and interpretation of these statistics requires an attention to detail which seems to tax the patience of working scientists. This high cognitive demand has led to an epidemic of shortcut definitions and i...
832 CitationsSource
#1Thomas R. Dyckman (FGCU: Florida Gulf Coast University)H-Index: 14
#2Stephen A. Zeff (FGCU: Florida Gulf Coast University)H-Index: 21
SYNOPSIS: This paper uses a sample of the regression and behavioral papers published in The Accounting Review and the Journal of Accounting Research from September 2012 through May 2013. We argue first that the current research results reported in empirical regression papers fail adequately to justify the time period adopted for the study. Second, we maintain that the statistical analyses used in these papers as well as in the behavioral papers have produced flawed results. We further maintain t...
42 CitationsSource
In this article we focus on three partial effect sizes for the correlation (r) family of effects: the standardized slope (b), the partial correlation (rp), and the semi-partial correlation (rsp). These partial effect sizes are useful for meta-analyses in two common situations: when primary studies reporting regression models do not report bivariate correlations, and when it is of specific interest to partial out the effects of other variables. We clarify the use of these three indices in the con...
40 CitationsSource
Cited By0