Typically, item analysis is used to check the test structure or questionnaires for internal consistency and reliability of the results. Cronbach's alpha is generated for item analysis and is usually referred to as a measure of internal consistency or reliability of the survey.
In this example, we use item analysis to compare the results of students' answers on a test paper. We are interested in investigating the correlation of the question results with each other and the consistency of the results.
The data is in the form of a short exam. 20 students are asked five questions. The results are 1 for a correct answer and 0 for an incorrect one.
The following steps will check the consistency of questions in a short test paper given to students:
- Open the
Item analysis.MTWworksheet. - Go to the Stat menu, click on Multivariate, and select Item Analysis….
- Enter the columns from
Q1toQ5in the Variables: section. - Click on OK to run the study.
Item analysis will display the correlation matrix and the matrix plot to look at the association between variables. A covariance matrix can also be displayed from the Results… option.
Cronbach's alpha is displayed along with an alpha table for one variable removed at a time. With the results observed here, removing Q2 from the study would increase alpha to 0.5797.
Cronbach's alpha is often referred to as a measure of internal consistency or reliability of the tests or questions. Values of 0.6 to 0.7 are thought to indicate a good level of consistency.