Why use multiple-choice questions on accounting - Bryant
Why use multiple-choice questions on accounting - Bryant
Why use multiple-choice questions on accounting - Bryant
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
<str<strong>on</strong>g>Why</str<strong>on</strong>g> Use Multiple Choice Questi<strong>on</strong>s 31reas<strong>on</strong> favoring CR tests is the greater likelihood of structural fidelity–i.e., the degree to whichexaminati<strong>on</strong> <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> require the same problem-solving skills encountered in the work venues ofa given field (Messick, 1993). This is particularly important in the <strong>accounting</strong> area, where employersare more interested in hiring competent accountants and auditors than good test takers.Despite their advantages, CR <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> have significant drawbacks, even for those whobelieve they are superior assessment tools. Perhaps the most important of them is that grading takesl<strong>on</strong>ger than for MC tests, tends to be more subjective, and often requires substantial prerequisiteknowledge. This process is also more <strong>on</strong>erous for the evaluators themselves, who are more likelyto be subject to both critical and litigious challenges. Finally, if the CR <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> require writingsamples or essays, some scholars (as well as many students) believe that CR <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> naturallyfavor those individuals with superior writing skills, even if poorly-written answers have superiorknowledge c<strong>on</strong>tent (Zimmerman and Williams, 2003).The MC-CR c<strong>on</strong>troversy includes <strong>on</strong>e final comp<strong>on</strong>ent: the questi<strong>on</strong> of “gender bias”(Walstad and Robs<strong>on</strong>, 1997; Hirschfeld et al.,1995). Certificati<strong>on</strong> test developers have a particularlystr<strong>on</strong>g interest in this matter beca<str<strong>on</strong>g>use</str<strong>on</strong>g> they have both a natural and a legal incentive to ensure thattheir examinati<strong>on</strong>s are “gender neutral”–i.e., that their tests do not favor males over females or viceversa. But are MC <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> really gender neutral?Given these c<strong>on</strong>cerns, the <str<strong>on</strong>g>choice</str<strong>on</strong>g> between using MC or CR <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> <strong>on</strong> a given professi<strong>on</strong>alcertificati<strong>on</strong> examinati<strong>on</strong> creates a natural dichotomy. Multiple-<str<strong>on</strong>g>choice</str<strong>on</strong>g> <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> are morec<strong>on</strong>venient to grade, but are c<strong>on</strong>sidered by many researchers to be less effective at measuring deepc<strong>on</strong>ceptual understanding (Becker and Johnst<strong>on</strong>, 1999), while CR <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> are just the opposite.The issues of “test equity” and “test efficacy” therefore come down to the extent to which the twotypes of <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> are related. If we can find a str<strong>on</strong>g relati<strong>on</strong>ship between them, then certificati<strong>on</strong>test developers can <str<strong>on</strong>g>use</str<strong>on</strong>g> MC examinati<strong>on</strong>s almost exclusively <strong>on</strong> such examinati<strong>on</strong>s, knowing thatwhatever is measured by CR tests is also measured by the other, and saving thousands of hours ofgrading time in the process. C<strong>on</strong>versely, if <strong>on</strong>ly a weak relati<strong>on</strong>ship exists–or n<strong>on</strong>e at all–thenexaminers would appear to be remiss in relying exclusively <strong>on</strong> MC <str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> in certificati<strong>on</strong>examinati<strong>on</strong>s. What empirical evidence exists to answer this questi<strong>on</strong>?Empirical EvidenceAn extensive body of research has addressed the questi<strong>on</strong> of how well MC versus CR<str<strong>on</strong>g>questi<strong>on</strong>s</str<strong>on</strong>g> test understanding of c<strong>on</strong>tent material. Much of the theoretical work comes from the areasof educati<strong>on</strong>al psychology and educati<strong>on</strong>al assessment (Martinez, 1999; Hancock, 1994; Nunnallyand Bernstein, 1994; Simkin and Kuechler, 2005). While these works suggest that it is theoreticallypossible to c<strong>on</strong>struct MC items that measure many of the same cognitive abilities as CR items, thequesti<strong>on</strong> remains of how well this theory holds up empirically.Early studies of this hypothesis have led some scholars to c<strong>on</strong>clude that MC tests and CRtests measure the same thing (Traub, 1993; Wainer and Thissen, 1993; Bennett et al., 1991;Bridgeman, 1991). After examining sample tests from seven different disciplines, for example,Wainer and Thissen (1993) c<strong>on</strong>cluded that this relati<strong>on</strong>ship was so str<strong>on</strong>g that “whatever is …measured by the c<strong>on</strong>structed resp<strong>on</strong>se secti<strong>on</strong> is measured better by the <str<strong>on</strong>g>multiple</str<strong>on</strong>g> <str<strong>on</strong>g>choice</str<strong>on</strong>g> secti<strong>on</strong>…Wehave never found any test that is composed of an objectively and a subjectively scored secti<strong>on</strong> forwhich this is not true” (p. 116). Subsequent studies by Walstad and Becker (1994) and Kennedy andWalstad (1997) echoed these sentiments.