Working With Data in SPSS Statistics

In this post, I will be using the following dataset:

What is Compute Variables?

Compute Variable is a tool that lets you make a new variable based on existing ones. It’s handy for tasks like combining or changing variables. Typically, it’s used to calculate averages or sums from questionnaire responses. For example, you can create a variable like “course satisfaction” by averaging participants’ answers to questions Q1 and Q2.

  • Name the new variable in the Target Variable box.
  • Provide a more descriptive Label for the new variable in the Type & Label dialogue.
  • Construct the Numeric Expression that will be utilized to compute values for the new variable. In this example, I am using this formula (Enjoy+Easy)/2.
  • By clicking the If button, you can access the If Cases dialogue, allowing you to define the cases to which the Numeric Expression should be applied.

As you can observe, the new variable for course satisfaction has been established.


Recode allows you to alter specific values or ranges of values on variables. This feature serves several purposes:

  • It enables collapsing continuous variables into categories. For instance, satisfaction scores could be categorized as “not satisfied” or “satisfied” by recoding scores of 3 or less as “1” (representing not satisfied) and scores higher than 3 as “2” (indicating satisfied).
  • Recode can combine or merge values, such as grouping courses into categories based on whether they require prior study of high-school human biology.
  • It facilitates reversing negatively scaled questionnaire items. This is useful when questionnaires contain both positively and negatively worded items, ensuring meaningful calculation of totals or averages by adjusting responses accordingly.
  • Recode can replace missing values within the data.
  • It helps in managing outliers and extreme scores by bringing them closer to the rest of the distribution, aiding in more accurate analysis.
  • The value for Category A is 1
  • The value for Category B is 2

Missing Value Analysis

In research or data collection, missing data can occur due to various reasons such as participants skipping questions, equipment failures, or spills. Dealing with missing data is crucial, and one method is using Expectation Maximization (EM) through Missing Value Analysis. This technique helps analyze the missing data pattern and provides options to impute values, filling in the gaps in the dataset.

Split File

Split File is a feature that lets researchers temporarily divide a data file into smaller groups based on shared characteristics. This allows for separate analysis of these subgroups, making it easier to compare them and draw conclusions.

Select Case

Frequently, researchers may prefer to analyze specific segments of a data file rather than the entire dataset. This objective can be accomplished through the utilization of Select Cases.


Frequencies are capable of generating summary statistics and simple graphs for a diverse array of variables.


Descriptives can calculate summary statistics that are typically applicable to interval and ratio variables. They can also determine a standardized score, known as a z-score, for each individual score within a dataset.


The Explore function can generate standard summary statistics, graphs, and tests for normality across multiple variables at once. These analyses can be conducted for the entire dataset or separately for each subgroup within it.

Chart Builder

The Chart Builder offers versatility in crafting various types of graphs.