Academic journal article College Student Journal

Effect of Test-Expectancy and Word Bank Availability on Test Performance

Academic journal article College Student Journal

Effect of Test-Expectancy and Word Bank Availability on Test Performance

Article excerpt

We examined test-expectancy as it applies to fill-in-the-blank tests. We randomly assigned 60 college students to take a fill-in-the-blank vocabulary test in one of three conditions. Two groups took the test with a word bank available; we told one group but not the other that they would have a word bank. The third group took the test with no word bank. There was no test expectancy effect; the two groups of students who took the test with the word bank did equally well. Students receiving a word bank outperformed students who did not have a word bank. The results suggest that informing students that they will have a word bank available may not have a detrimental effect on their performance.

**********

"Will the test be multiple choice or essay?" It is likely that millions of undergraduate students around the world ask this question of their instructors. Presumably students believe that if they know in advance what type of test to expect, they can be more prepared and in turn perform better. Test-expectancy, or knowing what type of test an instructor will administer, has been the subject of a growing amount of research beginning with studies conducted on the topic in the first half of the 20th century.

Meyer (1934) studied the effect of test-expectancy on subsequent test performance and found that students who expected a recall (essay) test performed better both on recall (essay) and recognition (multiple choice) tests than did those expecting a recognition (multiple choice) test. Since Meyer's original study, researchers have studied the effect of test-expectancy on test performance across a variety of situations. For instance, in a study involving word order and memory, D'Ydewalle (1981) showed that participants expecting a recall test recalled more words and more chunks (groups of words in sequence) than did those who expected a recognition test. Additionally, Oakhill and Davies (1991) showed that after reading a 500-word passage, participants expecting a recall test performed better on a recall test concerning the reading than did those expecting a recognition test. In addition, several other researchers have found results similar to those discussed, all agreeing that expecting a recall test contributes to increased test performance on both recall and recognition tests (Lundeberg & Fox, 1991; Neely & Balota, 1981; Sanjivamurthy & Kumar, 1983).

The differences in test performance based on test-expectancy may be related to differences in memory encoding processes. In a study employing wordlist memory tasks as a dependent measure, Neely and Balota (1981) found that those expecting a recall test performed better on both recall and recognition tests than did those expecting a recognition test. They proposed that these differences in test performance are facilitated by increases in the semantic organization of the to-be-remembered words on the part of the participant. Semantic organization refers to remembering words based on their meanings. The semantic organization effect was larger in recall-expectant groups, explaining the increased level of performance typical of recall-expectant participants. However, D'Ydewalle (1981) proposed that test-expectancy was facilitated not by how individuals encode information or store it in memory, but by how individuals retrieve information. These mixed findings may possibly be due to situational factors, thus future research could determine if test performance based on test-expectancy is due to encoding or retrieval.

Although results have historically shown support for the effect of test-expectancy on test performance, (i.e., superior performance of recall-expectant participants compared to recognition-expectant participants) several studies have reported results contrary to those previously discussed. Some studies have failed to show any effect of test-expectancy on recognition or recall (Hakstian, 1971; Kulhavy, Dyer, & Silver, 1975). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.