How to Calculate Z-Scores With the SPSS Student Version

By Matthew Lee

A z-score is a descriptive statistic used to determine how common or extreme a given score is by determining its distance from the mean in standard deviation units. Z-scores are calculated by subtracting the cell mean from actual scores, then dividing by the cell standard deviation. By converting actual scores to standardized z-scores (mean = 0, standard deviation = 1), this allows researchers to compare scores on scales with different units (e.g., weight in lbs vs. height in inches). Though SPSS does not readily provide z-scores in descriptive statistics tables, it is easy to produce and analyze z-scores by converting variables to standardized values.

Things You'll Need

  • SPSS Statistics Data Editor (Student Version)
  • Dataset

Converting Variables to Standardized Values

Step 1

Under the "File" menu, select "Open" then "Data" and open your data file. Once it loads, click on the "Analyze" menu, select "Descriptive Statistics" then "Descriptives."

Step 2

In the "Descriptives" window that appears, move your variables of interest to the "Variable(s)" column. You can select and analyze multiple variables at a time, and these can be moved to the "Variable(s)" column either by clicking and dragging or by highlighting the variables of interest and clicking on the arrow button.

Step 3

Click on the "Options" button and ensure that the "Mean" and "Std. Deviation" boxes are selected, then click on "Continue." Once out of the "Options" window, click on the "Save standardized values as variables" box at the bottom of the "Descriptives" window, then click on the "OK" button.

Step 4

Though you will automatically be taken to the "Output" window, your z-scores are not here. To find these, return to the window with your dataset. You will now have new columns (with "Z[variable name]" at the top), and the values in these columns are standardized z-scores.

Step 5

If you are interested in finding extreme scores or comparing scores on variables with different scales (among other things), analyze your newly created standardized variables just as you would the originals.