How to Run a Hierarchical Regression in SPSS

Written by linda foley
  • Share
  • Tweet
  • Share
  • Email

Social scientists use SPSS (Statistical Package for the Social Sciences) to analyse data. They use a hierarchical regression when they want to test the impact of specific predictor variables while controlling the influence of others. The hierarchical regression analysis allows the researcher to specify the order in which variables are entered into the procedure. The analysis tells the researcher how important a specific variable is in predicting a result.

Skill level:

Other People Are Reading

Things you need

  • SPSS
  • Data

Show MoreHide


    Analyse data

  1. 1

    Go to "Data View" on SPSS. Click on "Analyze" in the tool bar at the top of the page. Select "Regression" from the drop down menu and click on "Linear."

  2. 2

    When the dialogue box appears, move your dependent variable (for example, test score) into the "Dependent" box.

  3. 3

    Enter predictor variables, for example, sex, race and ses (socioeconomic status) into the "Independent" box. These are the variables you want to control.

  4. 4

    Click on "Next" which allows you to enter another variable or set of variables. Insert the variable(s) in which you are primarily interested, say "education level" in the "Independent" box. Click "OK."

    Read the Output

  1. 1

    Look at the first table "Variables Entered/Removed" which lists the variables you entered in steps 3 and 4 in the Analyze Data Section. Model 1 lists the variables you controlled (sex, race and ses). Model 2 is your variable of interest (education level).

  2. 2

    Look at the next table, the "Model Summary" which tells you the R Square for Model 1 with the variables you controlled (sex, race, ses) and Model 2, your variable of interest (education level).

  3. 3

    The important information is the change in R Square from Model 1 to Model 2. Subtracting the R Square of Model 1 (say .152 or 15.2%) from the R Square of Model 2 (say .303 or 30.3%) tells you how much predictive power your variable of interest has. In this case, 30.3% - 15.2% = 15.1%, which means your variable predicts 15.1% of the difference.

Don't Miss

  • All types
  • Articles
  • Slideshows
  • Videos
  • Most relevant
  • Most popular
  • Most recent

No articles available

No slideshows available

No videos available

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