How to Enter Multiple Responses in SPSS: Complete Guide with Examples

Multiple response questions allow survey respondents to select more than one answer from a list of options. Analyzing these questions in SPSS requires a specific approach because each respondent can have multiple selections, unlike single-choice questions where only one answer is possible.

This guide shows you how to properly enter, code, and analyze multiple response data in SPSS, from initial data preparation in Excel through running frequency and crosstab analyses.

What Are Multiple Response Questions?

Multiple response questions (also called "check all that apply" or "select all that apply") present respondents with a list of options where they can choose one or more answers. These questions are common in surveys for measuring preferences, behaviors, or characteristics.

Example multiple response question:

"Which of the following sports shoe brands have you purchased in the past year? (Select all that apply)"

Unlike single-choice questions where each respondent selects exactly one option, multiple response questions generate complex data where each respondent may select none, one, or several options.

Understanding Multiple Response Data Structure

Before entering data into SPSS, you need to understand how multiple response data must be structured. There are two common approaches:

Dichotomous Coding (Recommended)

Each response option becomes a separate variable coded as:

• 1 = Option selected
• 0 = Option not selected

For our sports shoe example with 6 brands, you would create 6 separate variables (Adidas, Nike, Clarks, Lotto, Fila, Puma). If a respondent selected Adidas and Nike, their data would be:

This method is preferred because it clearly represents each option's selection status and works seamlessly with SPSS multiple response analysis.

Category Coding (Alternative)

Each selection gets numbered sequentially in separate columns. If a respondent selected Adidas (coded as 1), Nike (coded as 2), and Fila (coded as 5), their data would be:

While this method saves space, it's less intuitive and requires additional coding. We recommend the dichotomous approach for clarity and ease of analysis.

Step 1: Prepare Data in Excel

Proper data preparation in Excel is critical for successful SPSS analysis. Follow these steps carefully:

Create Your Data Structure

1. Open a new Excel spreadsheet
2. Create column headers for each response option using clear, descriptive names without spaces (use underscores instead)
3. Add a respondent ID column as the first column for tracking

Example spreadsheet structure:

Code Your Responses

For each respondent:

• Enter 1 if they selected the option
• Enter 0 if they did not select the option
• Use 0 for all options if the respondent skipped the question (or use a missing value code if preferred)

Important coding rules:

• Be consistent with your coding scheme
• Avoid leaving cells blank (use 0 or a missing value code)
• Double-check data entry for accuracy
• Keep variable names short but descriptive (max 64 characters)

Save Your File

1. Save the Excel file in a location you can easily access
2. Use a descriptive filename (e.g., "Survey_MultipleResponse_Brands.xlsx")
3. Remember the file path for importing into SPSS

Step 2: Import Data into SPSS

Once your Excel file is properly formatted, import it into SPSS:

1. Open SPSS
2. Go to File → Import Data → Excel
3. Navigate to your Excel file and click Open
4. In the import wizard:
   • Check "Read variable names from the first row of data" if your first row contains headers
   • Verify the data preview looks correct
   • Click OK

Your data should now appear in SPSS Data View with all variables and cases properly imported.

Step 3: Assign Value Labels

Value labels make your output more readable by displaying meaningful text instead of numbers.

1. Switch to Variable View (click the "Variable View" tab at the bottom)
2. Locate the Values column for your first multiple response variable (e.g., Adidas)
3. Click in the Values cell to open the Value Labels dialog
4. Add labels:
   • Value: 0, Label: "Not selected" (or "No")
   • Value: 1, Label: "Selected" (or "Yes")
5. Click OK

Copy Labels to Other Variables

Rather than manually entering labels for each variable, copy them:

1. Click in the Values cell of the variable you just labeled
2. Copy (Ctrl+C or Cmd+C)
3. Select the Values cells for all other multiple response variables
4. Paste (Ctrl+V or Cmd+V)

This saves time and ensures consistency across all variables.

Step 4: Define Multiple Response Sets

SPSS treats multiple response questions as a set of related variables. You must define this relationship before analysis.

1. Go to Analyze → Multiple Response → Define Variable Sets
2. In the Define Multiple Response Sets dialog:
   • Select variables: Highlight all variables in your multiple response set (e.g., Adidas, Nike, Clarks, Lotto, Fila, Puma) and move them to the "Variables in Set" box
   • Variables are coded as: Select "Dichotomies" (since we're using 1/0 coding)
   • Counted value: Enter 1 (this tells SPSS that 1 means "selected")
   • Name: Enter a short name for your set (e.g., "$brands" - the$ prefix is standard)
   • Label: Enter a descriptive label (e.g., "Sports Shoe Brands Purchased")
3. Click Add to create the set
4. Click Close

Your multiple response set is now defined and ready for analysis.

Step 5: Run Frequency Analysis

Frequency analysis shows how many respondents selected each option and the percentage of total responses.

1. Go to Analyze → Multiple Response → Frequencies
2. In the Frequencies dialog:
   • Move your multiple response set (e.g., $brands) from the left box to the "Table(s) for" box
   • (Optional) Adjust table formatting preferences
3. Click OK

Interpreting Frequency Output

The output table shows two important columns:

• Responses:

  • N: Number of times each option was selected
  • Percent: Percentage of all responses (may total > 100% since respondents can select multiple options)

• Cases:

  • N: Same as Responses N
  • Percent: Percentage of total respondents who selected this option

Example interpretation: If your output shows Nike was selected by 65 respondents out of 100 total respondents, the "Cases Percent" would be 65%. This means 65% of respondents purchased Nike shoes in the past year.

Step 6: Run Crosstab Analysis

Crosstabs allow you to examine relationships between multiple response questions and other categorical variables (e.g., comparing brand preferences by gender, age group, or region).

1. Go to Analyze → Multiple Response → Crosstabs
2. In the Crosstabs dialog:
   • Rows: Move your multiple response set (e.g., $brands)
   • Columns: Move your categorical variable (e.g., Gender)
   • (Optional) Check "Display clustered bar charts" for visualizations
   • (Optional) Click "Options" to customize percentages shown
3. Click OK

Interpreting Crosstab Output

The crosstab table displays your multiple response options in rows and your categorical variable categories in columns. Each cell shows:

• Count of respondents who selected that option within that category
• Percentage (based on options selected in the "Options" dialog)

Example interpretation: If your crosstab shows that 40 out of 50 male respondents (80%) selected Nike, while only 25 out of 50 female respondents (50%) selected Nike, this suggests Nike has stronger appeal among male respondents in your sample.

Practical Example: Sports Shoe Brand Preferences

Let's walk through a complete example with 20 respondents answering: "Which sports shoe brands have you purchased?"

Sample Data

Analysis Steps

1. Import data from Excel into SPSS
2. Assign value labels: 0 = "Not selected", 1 = "Selected"
3. Define set: Create $brands set with all 6 brand variables
4. Run frequencies: Analyze → Multiple Response → Frequencies
5. Run crosstabs: Cross-tabulate brands by demographic variables

Expected Findings

Frequency analysis would reveal:

• Which brands are most/least popular overall
• Total number of brands selected per respondent (on average)
• Percentage of respondents selecting each brand

Crosstab analysis would show:

• Brand preferences by gender, age, income level
• Patterns in multiple brand selection
• Demographic segments with highest brand loyalty

Common Issues and Solutions

Issue: "Variables must be coded as dichotomies" Error

Cause: Your variables contain values other than 0 and 1, or SPSS doesn't recognize your coding scheme

Solution:

• Verify all cells contain only 0 or 1 (no blank cells, text, or other numbers)
• In Define Multiple Response Sets, confirm you selected "Dichotomies" and entered "1" as the counted value
• Check for data entry errors in Variable View

Issue: Percentages Total More Than 100%

Explanation: This is normal for multiple response analysis! Since respondents can select multiple options, the sum of percentages across all options typically exceeds 100%

Example: If 80% of respondents selected Nike and 70% selected Adidas, the total is 150%. This simply means many respondents selected both brands.

Issue: Missing Values Affect Results

Cause: Blank cells or inconsistent missing value codes distort frequency counts

Solution:

• Replace all blank cells with 0 (if no selection) or a consistent missing value code (e.g., 999)
• In SPSS Variable View, define missing values if using a specific code
• Recheck data preparation in Excel before importing

Issue: Variable Names Appear Instead of Labels

Cause: Value labels weren't assigned or copied correctly

Solution:

• Return to Variable View
• Verify each variable has value labels (0 = "Not selected", 1 = "Selected")
• If labels are missing, add them and copy across all variables
• Ensure "Display value labels" is enabled in SPSS preferences

Advanced Techniques

Creating Summary Variables

You can create a new variable that counts total selections per respondent:

1. Go to Transform → Compute Variable
2. Target Variable: Enter a name (e.g., "Total_Brands")
3. Numeric Expression: Enter: Adidas + Nike + Clarks + Lotto + Fila + Puma
4. Click OK

This creates a variable showing how many total brands each respondent selected (range: 0-6 in our example).

Filtering Data for Specific Responses

To analyze only respondents who selected a particular option:

1. Go to Data → Select Cases
2. Choose "If condition is satisfied" and click If
3. Enter your condition (e.g., Nike = 1 for respondents who selected Nike)
4. Click Continue then OK

Your subsequent analyses will include only respondents who meet this condition.

Comparing Multiple Response Sets

If you have multiple sets of multiple response questions (e.g., brands purchased vs. brands considered), you can compare them:

1. Define separate multiple response sets for each question
2. Run crosstabs with one set in rows and the other in columns
3. Analyze overlap between the two sets

Best Practices for Multiple Response Analysis

Data Preparation

• Use consistent coding: Always use 1 for selected and 0 for not selected
• Create clear variable names: Use descriptive names without spaces (e.g., Brand_Adidas)
• Include all options: Even rarely selected options should have their own variable
• Document your coding scheme: Keep notes on what each variable represents

Analysis Strategy

• Start with frequencies: Understand overall response patterns before crosstabs
• Check sample sizes: Ensure sufficient respondents per category for meaningful comparisons
• Consider response rates: Low selection rates (< 5%) may indicate unclear options or survey design issues
• Validate patterns: Use multiple analysis techniques to confirm findings

Reporting Results

• Clarify percentage type: Always specify whether percentages are based on responses or cases
• Report sample size: Include total respondents and total selections
• Use visualizations: Bar charts and clustered bar charts make patterns easier to identify
• Provide context: Explain why percentages may exceed 100% for readers unfamiliar with multiple response data

Limitations of Multiple Response Analysis in SPSS

While SPSS handles multiple response data effectively, there are important limitations:

Statistical Tests Not Available

SPSS does not support inferential statistics (chi-square, t-tests, etc.) directly on multiple response sets. You can only perform:

• Frequency analysis
• Crosstab analysis

For statistical tests, you must analyze individual dichotomous variables separately or use specialized procedures.

Complex Crosstabs Restricted

You cannot include multiple response sets in three-way or higher-dimensional crosstabs within the Multiple Response module. For complex analyses, consider:

• Breaking down into multiple two-way crosstabs
• Using custom tables (Analyze → Tables → Custom Tables) if available
• Analyzing individual variables separately

Missing Data Handling

SPSS treats missing values in multiple response sets differently than in standard variables. Carefully plan how to handle:

• Respondents who skip the entire question
• Partial responses (if using category coding)
• "None of the above" or "Not applicable" responses

Wrapping Up

Entering and analyzing multiple response questions in SPSS requires proper data preparation and understanding of the multiple response module. By following the dichotomous coding approach (1 for selected, 0 for not selected), defining response sets correctly, and using frequency and crosstab analyses, you can effectively examine patterns in data where respondents select multiple options.

Remember that percentages exceeding 100% are normal in multiple response analysis because respondents can choose several options. Always prepare your data carefully in Excel before importing to SPSS, assign clear value labels for readability, and document your coding scheme for reference.

While SPSS limits you to descriptive analyses (frequencies and crosstabs) for multiple response sets, these tools provide powerful insights into response patterns, option popularity, and relationships with demographic or behavioral variables.

References

• IBM SPSS Statistics. (2024). Multiple Response Analysis. IBM Documentation.
• Field, A. (2024). Discovering Statistics Using IBM SPSS Statistics (6th ed.). SAGE Publications.
• Pallant, J. (2023). SPSS Survival Manual (7th ed.). Routledge.