skip to main content skip to footer

Creating Better Classroom Tests

Author(s):
Gulliksen, Harold O.
Publication Year:
1985
Report Number:
RM-85-05
Source:
ETS Research Memorandum
Document Type:
Report
Page Count:
72
Subject/Key Words:
Classroom Tests, Educational Methods, Test Construction

Abstract

There has been no improvement, perhaps even a decline, in the quality of teacher-made, small classroom tests over the last 40 years. Many other types of objective items are better than multiple-choice tests. There are many types of free-answer items other than essay tests. A handbook on item writing which could be used by teachers interested in writing better items for their tests was prepared by Sybil Carlson of the Educational Testing Service. It is called "Creative Classroom Testing: Ten Designs for Assessment and Instruction." About 1930, the President of the University of Chicago introduced an examination system. The curriculum for the first and second years was five one-year courses: 1) biological science; 2) English; 3) humanities: 4) social studies; and 5) physical science. Passing each course consisted of passing a six-hour exam. The examiners and teaching faculty at the University of Chicago constructed the test items together; they were then reviewed, revised, and approved by the course's teaching staff. There were one- or two-page essays; short-answer items; and various objective items. The exams at Chicago were not entirely written. Where laboratory work was important, the exam involved laboratory setups. Part of teacher training in schools of education should be writing items of these various types. Teachers should write items over a period of time. The should give a free-answer exam, then use the student's wrong answers for distractors. During World War II, the author and others were on a project for the Navy to develop achievement and aptitude tests for the Navy schools. Aptitude tests were very useful in assessing the content validity of achievement tests. Investigating the grading procedure and testing showed that important aspects of the work were inadequately tested, but less important parts were very important in grading. Aptitude tests can indicate whether course grades are measuring relevant or unimportant parts of the course. Multiple-choice should be used only when it is appropriate. The teacher should use the item type that is most convenient and natural for the subject. Performance and identification items are useful.

Read More