How to test for normality in SPSS

Written by stephen byron cooper Google
  • Share
  • Tweet
  • Share
  • Pin
  • Email
How to test for normality in SPSS
SPSS is produced by IBM. (Getty Thinkstock)

SPSS is a statistical analysis computer application produced by IBM. The SPSS Statistics module of the package enables analysis of groups of observations. A normality test checks whether findings plot into a “normal distribution,” which is usually represented by a “bell curve.” The bell curve is the classic shape of a plot of population data. Once you have your data Entered into the system, you can test several variables simultaneously.

Skill level:
Moderately Easy

Other People Are Reading

Instructions

  1. 1

    Prepare your data in SPSS, putting observations for each variable in a column. Each column should have a title. You need to identify the variables by name when performing the normality test.

  2. 2

    Click on “Analyse” in the menu bar and go to “Descriptive Statistics” in the drop-down menu. Select “Explore” in the sub-menu. This opens up the “Explore” window.

  3. 3

    Select the variables you want to test from the list in the left panel. Click on each, then press the top arrow button in the middle of the screen to populate the “Dependent List” box.

  4. 4

    Click the “Plots” button to open the “Explore: Plots” window. Make sure the “Normality plots with tests” and “Histogram” check boxes have ticks in them. Press the “Continue” button to close the “Explore: Plots” window.

  5. 5

    Press the “OK” button to start the test. The “Explore” window will close and the Explore results will be shown in the main panel of SPSS.

  6. 6

    Scroll down through the results to get to a table with the title “Tests of Normality.” This will show two tests. The first test of normality is called the Kolmogorov-Smirnov test. This is not a very reliable test, so don’t use this. Look at the second test, which is the Shapiro-Wilk test. This gives a more accurate reading.

  7. 7

    Read the “Sig” figure in the Shapiro-Wilk test. If this number is less than 0.05 then the dataset does not pass the test of normality.

Don't Miss

Filter:
  • All types
  • Articles
  • Slideshows
  • Videos
Sort:
  • Most relevant
  • Most popular
  • Most recent

No articles available

No slideshows available

No videos available

By using the eHow.co.uk site, you consent to the use of cookies. For more information, please see our Cookie policy.