Item Difficulty and Programmed Learning

Author(s):

Jacobs, Paul I.

Publication Year:

1961

Report Number:

RM-61-05

Source:

ETS Research Memorandum

Document Type:

Report

Page Count:

35

Subject/Key Words:

United States Air Force, Adaptive Testing, Difficulty Level, Programed Instruction, Test Construction, Test Items

Abstract

Item difficulty, as traditionally conceived by test developers, is response-defined, is a group measure, and is sensitive to the conditions under which it is made. A major principle of programmed instruction is that all items should be answered correctly by all or most students. In fact, it is desirable that the programmer uses students' responses as feedback and tries to eliminate student error during training. This information may also be used to determine what material the learner should be exposed to next. Many different manipulations can influence item difficulty. However, research indicates that these manipulations do not affect learning in the same way. Thus, there is no empirical justification for desiring low item difficulty per se. The writer of programmed learning might consider those techniques which reduce error and simultaneously enhance learning, such as the use of prompts and individualizing the programs according to individual aptitude. Further attention to individual ratings of item difficulty and/or response latency, may be very useful.

Read More

• Request Copy (specify title and report number, if any)