SPSS manual: practical tips for using SPSS

Use a codebook for your questionnaire

In your code book, which typically comprises several A4 pages, you record the responses to your questionnaire or survey using codes and specify their meanings. For instance, for ‘gender,’ you assign the code 1 for male, 2 for female, and 3 for neutral. To maintain clear organization, provide a well-structured table, as exemplified below, that displays the question number, the corresponding question, and the associated code(s). It is possible for a question number and question to have multiple codes assigned to them. However, it is common practice to use the code 0 solely for dummy variables; for all other cases, it is advisable to commence the coding process from 1.

Processing your codes in SPSS

You process your codes in the Variable View. You can fill in the following fields here:

• Name: You can enter the question number without spaces, for example V1, V2, etc. However, you can also give a variable a name, for example ‘gender’. This often makes it easier to search for the correct variable when performing your data analyses, especially if you have a dataset with many variables.
• Type: Here you enter the type of data; numeric for data in numbers and string for data in letters.
• Width: Here you can enter the width of the column, which is often only necessary for string data because more characters need to be accommodated. You can increase the column number.
• Decimals: Here you can indicate the number of digits after the decimal point, which is only useful for numerical data. By default, the number of decimal places is set to 2.
• Label: Here you enter the name of the variable, which is the second column of your codebook, such as gender.
• Values: Here you enter the codes from the third column. For example, for gender, you would enter 1 = male, 2 = female, etc.
• Missing: Here you indicate how SPSS should handle missing data, for example, if a respondent skips a question. You can also assign a code for this. It is common to use codes 9, 99, or 999 for missing data. However, if you have a question with 9 (or more) answer categories, you cannot use code 9 for missing values because it is a valid code for that question (or questions). In that case, you would use 99 or 999.
• Columns: Here you can display the width of the column as you see it in your data screen. You can adjust the width if needed, but typically this is not necessary.
• Align: You can choose from three options here: left, right, and center. SPSS defaults to aligning numeric variables to the right and string variables to the left.
• Measure: You have three options here: nominal, ordinal, and scale. This indicates the level of measurement for your variables. ‘Gender’ in the table above is a nominal variable, education level is an ordinal variable, and the questions with statements are scale variables.
• Role: This allows you to indicate the role a variable will play in your analysis. Is it an independent (input) or dependent (target) variable, or both? SPSS automatically sets variables as input. It is usually not necessary to adjust this for your analyses.

Enter data

The next step is to input the data and verify for any missing information. You can enter the data in the Data View screen. There are numerous videos available on YouTube that provide detailed instructions on how to do this accurately.

Analysis of the data, descriptive statistics

Your data analysis begins with describing the data. This can be accomplished through frequency tables and measures of central tendency and variability. Frequency tables are particularly useful for nominal and ordinal variables. To request and generate frequency tables, you can follow these steps in the menu: Go to Analyze, then Descriptive Statistics, and finally Frequencies.

In the video below, you will be able to see how to do this instantly.

SPSS then gives the following output (example):

Sex

For a continuous variable, you typically request the mean and standard deviation. To obtain these values, you can follow these steps: Go to Analyze, then select Descriptive Statistics and Descriptives. Alternatively, you can choose Analyze, then Compare Means, and finally Means. This is well explained in the next video.

Scale variables include age, as well as questions where respondents are required to indicate the degree to which they agree with a certain statement.

Age

Of course, you can also request a frequency distribution for scale variables. For example, if you are interested in knowing the number and percentage of respondents who answered “completely disagree,” “disagree,” and so on.

Look for patterns in the crosstabs

After you have described the data, you can begin your research and identify the prominent patterns. One useful method to uncover these patterns is through the use of crosstabs. Here, you compare two variables. Cross-tabulation is particularly suitable for analyzing nominal and ordinal variables. To obtain a crosstab in SPSS, you can follow these steps: Go to Analyze, select Descriptive Statistics, and click on Crosstabs.

Would you like more explanation about how to extract the patterns from the crosstabs? This is explained in the video below.

Do you want help with your SPSS analyses?

In this article, we provide you with general tips about SPSS. Would you like to discuss the best way to set up your research or determine which variables to include? We are here to help. Do you need assistance with loading your dataset, creating a codebook, or conducting the analyses?

We are here to assist you with following SPSS analyses:

• Correlations and descriptive statistics
• Reliability and T-tests (t-test)
• Chi-square test, Wilcoxon Signed-Ranks test, Mann-Whitney test (non-parametric)
• Repeated and Within Measures analyses
• ANOVA, ANCOVA and MANOVA
• Linear Regression, Logistic Regression and OLS regression
• Kruskal-Wallis test
• Cluster Analysis, Factor Analysis, Power Analysis
• Panel analysis/ multilevel analysis/ mixed models
• Conjoint analysis and time series analysis