Changing Initial Answers on Multiple-Choice Achievement Tests

Pascale, Pietro J.

doi:10.1080/00256307.1974.12022604

Cited by 22 publications

(15 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If one were to view extensive answer changing as a sign of test-taking uncertainty, then it is not surprising to find that most authors report an inverse relationship between test scores and the numbers of revisions made by students: Six studies reported a statistically significant negative relationship (Lynch & Smith, 1975;Mueller & Shwedel, 1975;Stoffer et aI., 1977;Johnston, 1978;Best. 1979;Sitton, Adams & Anderson, 1980) but five others reported nonsignificant results in the same direction (Reile & Briggs, 1952;Reiling & Taylor, 1972;Pascale, 1974;Smith & Moore, 1976;Crocker & Benson, 1980). Only in one study is there a positive relationship (Beck, 1978) and it was nonsignificant, and two other studies showed no trend in either direction (Bath, 1967;Copeland, 1972).…”

mentioning

confidence: 85%

“…Interestingly, which side of the fence these strategists were on was often not clear. For example, Huff's (1961) position was both pro and con, depending on the article in which he was cited (Jacobs, 1972;Pascale, 1974;Davis, 1975;Lynch & Smith, 1975;Stoffer, Davis, & Brown, 1977). For the record, Huff's position is a qualified "yes" to answer changing, except in "a case that is very close to sheer guess" (p, 36).…”

mentioning

confidence: 93%

“…Achievement variables include course grade and grade point average (G.PA), whereas scores on Borgatta and Corsini's (1964) Quick Word Test (QWT), the Scholastic Aptitude Test (SAT), and the Airman Qualifying Exam (AQE) have been used as ability variables. Individuals who were rated low on these variables typically were found to change more answers (Mathews, 1929;Reiling & Taylor, 1972;Pascale, 1974;Stoffer, et aI., 1977), however, in only the last of those studies was there found a significant relationship between ability (as measured by the SAT) and frequency of answer changing. In none of these studies was there any relationship between any of the achievement/ability variables and the degree of profit from answer changing.…”

mentioning

confidence: 94%

“…The statement that increased answer changing will improve one's score is tempered, not only by the data on test scores, but also by the conflicting data on net gains enjoyed versus the numbers of revisions made: Six studies have shown that profit from answer changing increased significantly with the number of revisions (Reiling & Taylor, 1972;Pascale, 1974;Mueller & Shwedel, 1975;Stoffer et aI., 1977;Sitton et aI., 1980;Range, Anderson & Wesley, 1982), three others found negative but nonsignificant trends (Lehman, 1928;Mathews, 1929;Archer & Pippert, 1962), and one study showed no relationship in either direction (Berrien, 1939). Obviously a point must be reached where answer changes are no longer beneficial to the test-taking student.…”

mentioning

confidence: 97%

“…Sex has also been investigated as a subject variable in predicting answer changing frequency and/or gains in a number of studies (ReiIe & Briggs, 1952;Mallinson & Miller, 1956;Bath, 1967;Copeland, 1972;Reiling & Taylor, 1972;Pascale, 1974;Mueller & Shwedel, 1975;Stoffer, et aI., 1977;Beck, 1978;Sitton, et aI., 1980;Skinner, 1983). Three of these studies showed a slightly higher frequency of answer changing by females and one study (Skinner, 1983) reported the only statistically significant sex difference, finding that females changed twice as many answers as males.…”

mentioning

confidence: 98%

See 4 more Smart Citations

Staying with Initial Answers on Objective Tests: Is it a Myth?

Benjamin

Cavell

Shallenberger

1984

Teaching of Psychology

View full text Add to dashboard Cite

The common advice to not change answers appears to be a mistake, but before that is certain, additional information is needed.Since 1928, at least 33 studies have been published concerning a number of issues surrounding answer-changing behavior on objective tests. Although results in these studies have sometimes been at variance, the one consistent finding is that there is nothing inherently wrong with changing initial answers on objective tests. In fact, the evidence uniformly indicates that: (a) the majority of answer changes are from incorrect to correct and (b) most students who change their answers improve their test scores. None of the 33 studies contradicts either of those conclusions.Most of the research in this area has been aimed at testing the accuracy of "first impressions" in test-taking. This bit of academic folk wisdom is typically stated as the belief that one should not change answers on objective tests because initial reactions to test questions are intuitively more accurate than subsequent responses.Prevalence of the Belief and Potential Sources. That this belief is widespread among test-takers is supported by a number of studies. Surveys of students' attitudes toward the results of answer changing are fairly consistent in their outcomes revealing that most students (between 68% and 100%) do not expect changed answers to improve their score. Indeed, approximately three out of every four of these students felt answer changes would lower their score. The percentages of students reporting an expected improvement have ranged from 0%

show abstract