A LINE ON LIFE

11/28/84, Updated 11/5/01

Objective Tests

Should You Stay with Your First Answers?

David A. Gershaw, Ph.D.

Academic "folk wisdom" has emphasized the accuracy of "first impressions." In other words, you typically should not change initial answers on an objective (true false or multiple choice) test. If you do, you are more likely to change to a wrong answer. Since we take objective tests even after we are out of school (driver's license or civil service exam), it is important for us to know is this myth really true?

A psychological article* summarized 33 studies that were conducted from 1928 to 1983. Although there are some differences in the findings of these studies, the evidence gives two uniform indications:

1. The majority of answer changes are from incorrect to correct.

2. Most students who change their answers improve their test scores.

Do students think that changing their answers will help their scores? In contrast to what actually occurs, between 68% and 100% do not expect changed answers to improve their scores. In fact, about 3 out of every 4 students believed that answer changes would lower their scores.

Why do students believe this? One possibility is that the myth is reinforced by their instructors. To test this hypothesis, a study was done of the Education, Liberal Arts and Science faculty at Texas A & M University. Of those faculty who used objective tests, most of them (55%) believed that changing initial answers would lower student scores. Only about 15% thought it would improve student scores. (The remainder either indicated "no change" or "don't know.")

About one-third of these instructors indicated that they gave their students instructions about changing answers. Of those who gave instructions, almost two-thirds (63%) warned students not to change answers, because they would get more wrong answers!

What proportion of students actually change their answers? Among the studies reviewed, anywhere from 57% to 96% changed their answers, with a median of 84%. What proportion of items were changed? Although the percentages ranged from 2% to 9%, it was typically just above 3%.

Did these changes hurt or help? On true-false items, you can change from "wrong to right" or "right to wrong." However, with multiple-choice items, you can also change from one wrong answer to another. Below are the average (median) percentages for each category.

CHANGES

Multiple-Choice Studies (20 studies)

Wrong to Right 57.8%

Right to Wrong 20.2%

Wrong to Wrong 22.8%

True-False Studies (5 studies)

Wrong to Right 66.6%

Right to Wrong 33.5%

Thus in spite of the myth about changing answers the ratio of gainers (wrong to right) to losers (right to wrong) is 2/1 for true-false tests and almost 3/1 for multiple choice. You are 2-3 times more likely to get it right, if you make a change!

If this is so, why do students still believe the myth? One reason is that they have not received information to the contrary, (That is why I am writing this article.) Another possibility is selective perception. In other studies, students seldom remember items they changed to the correct response. In contrast, they seem to "highlight" in their memory those items that they change and get wrong.

However, it is not quite that simple. The success of your answer may depend on your reason for the change. If you mistakenly put down the wrong answer or misread the question (like missing the word, "not") you are very likely to increase your score by changing your answer. However, if you are just unsure of your answer or merely guessing for you, the myth may be no myth at all.


* Ludy T. Benjamin, Timothy A. Cavell and William R. Shallenberger III, "Staying with Initial Answers on Objective Tests" Is It a Myth?" Teaching of Psychology, October, 1984, pages 133-141.

Go back to listing of additional articles.

Go back to "A Line on Life" main page.