Cronbach's alpha using SPSS: An approach to Reliability Analysis

Cronbach's alpha using SPSS

There are several domains of reliability analysis, one of them is internal consistency.

Steps of computing Cronbach's Alpha using SPSS and how to interpret the result?

Step 1: Open dataset. Here I have used sample dataset "demo.sav" which is in installation directory. After that determine the list of variables that measure a particular construct. Usually, we compute Cronbach's alpha for likert scale type questions. However, we can also compute for yes/no type questions.

Step 2: Go to analyze, scale and click on "reliability analysis".

Step 3: Then the following dialog box will open. Drag the questions to the right in the "item" box for which you want to calculate Cronbach's Alpha. Here, the example variables measure the construct "Means of communication", so I dragged the questions related to means of communications to the right box such as wireless, multiline, voice mail, pager, internet and so on. In the model, select "Alpha" as shown in the figure below. And then click on "Statistics".

Step 4: When you click on Statistics, the following dialog box appears. Then select the "Descriptives for" options and choose "correlations" under inter-item heading. Then click continue.

Step 5. Click on okay

Step 6: Then the following output is displayed.

Interpretation of output

You can see that the Cronbach's alpha value for 14 items is shown to be approximately 0.61. The cutoff value of 0.7 is usually used in social science researches. So, Cronbach's value of 0.7 or higher is generally considered reliable. Value ranges from 0 to 1.

Also, you can see a table in which you see how much cronbach's alpha value will change if you delete an item from the scale. Although overall cronbach's alpha was 0.61, it would increase if you would delete one item "Newspaper subscription" from the model as shown in figure below.

So, if you re-run the overall step by removing that variable "Newspaper subscription", then there will be change in the output. i.e. value increased from 0.61 to 0.66. This is because it was re-calculated for 13 items. So, we keep on following the same steps to remove unreliable items and make our tool reliable. However Cronbach's alpha should't be used as a sole criteria for measuring reliability. It is just a statistical guide for quick decisions on reliability measures.

Here is a video tutorial on the detailed steps:

Thank you and see you in the next tutorial.