Testing the validity of the questionnaire as a measuring tool can be conducted in various ways. The ideal is to use Confirmatory Factor Analysis (CFA). There are several software programs that can be used for CFA, such as LISREL, AMOS, and R. However, in the practice of writing a thesis, another method is frequently performed (simpler but less stringent tests), which is called the simple correlation technique, so this can be conducted by SPSS software, easily.
Example of Testing the Validity
For example, the operationalization of the variables is as follows.
Table 1
Suppose the questionnaire uses a Likert scale (1 = “Strongly Disagree” to 5 = “Strongly Agree”), and 20 of the 100 responses to each question item are as follows.
Table 2
(Complete responses can be downloaded at the following link: (click here)
Validity testing is conducted per dimension or subvariable. For example, we will check whether items No. 1 to 4 are valid for measuring “OPENNESS”.. Before inputting data into SPSS, add a column to the right of column “No. 4”. Fill in the column with the total score of the responses for items No. 1 to No. 4. The results are represented by the following table.
Table 3
Enter all the data points (100 rows, each with 5 columns) into the SPSS spreadsheet. See the table below.
Table 4
In Table 4, opn1, opn2, opn3, opn4, and sum_opn are the new names for No. 1, No. 2, No. 3, No. 4, and SUM in Table 3. To get the data in .sav, click here. Below is the SPSS output that shows Spearman’s correlation coefficients among the variables.
Table 5
As shown in Table 5, the Spearman’s correlation coefficients between each item and the Sum of the items’ scores are 0.782, 0.864, 0.712, and 0.697. The output also shows that all the correlations are significant. Based on these, we can conclude that all the items are valid for measuring “OPENNESS”.
