Statistics for Research - L5 - Import Data from Google Forms into SPSS
Based on Research With Fawad's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.
Convert Google Forms response text (e.g., “strongly agree”) into the questionnaire’s numeric codes using Google Sheets “Find and replace.”
Briefing
Importing Google Forms responses into SPSS starts with cleaning the exported data so SPSS can treat survey items as numeric variables instead of text. After opening Google Sheets and switching to “View in Sheets,” the responses appear as text labels like “strongly agree,” “strongly disagree,” “disagree,” “agree,” “neutral,” plus categorical fields such as gender. The workflow uses Google Sheets’ “Edit → Find and replace” to convert each Likert option into the numeric codes used in the questionnaire (e.g., “strongly agree” becomes 5, “strongly disagree” becomes 1, “disagree” becomes 2, “agree” becomes 4, and “neutral” becomes 3). A key detail is avoiding accidental partial matches: “agree” must be replaced with “match entire cell contents” because “agree” appears inside “strongly disagree” and “strongly agree.”
The same replacement approach handles other categorical variables. For gender, the transcript maps “male” to 1 and “female” to 2, but it highlights a common pitfall: if “match entire cell contents” isn’t used consistently, earlier replacements can shift values (e.g., “female” may become “fee 1” after “male” is replaced). The fix is to correct the resulting codes in a follow-up replacement so the final mapping is consistent (female ends up as 2). Once all text-to-number conversions are complete, the cleaned dataset is downloaded from Google Sheets as either XLS or CSV.
Before importing into SPSS, the XLS version may require renaming variables because SPSS can struggle with variable names that include spaces. The transcript demonstrates manually renaming columns in the spreadsheet (e.g., conflict items become CF1, CF2, etc.; community items become CM1, CM2; leadership items become leadership-style labels; team effectiveness items; and exhaustion items like ex1, ex2). It also removes unnecessary columns such as timestamps to keep the dataset focused.
With the XLS file ready, SPSS imports it via “File → Import Excel,” using options like “read variable names from first row of the data.” After import, the data view shows variables correctly populated, enabling downstream statistical analysis.
CSV import follows the same logic but with extra caution about formatting. The transcript notes that if text contains commas, CSV can break fields, so it recommends ensuring the variable naming row is correct and removing the timestamp column if needed. In SPSS, CSV is imported through “File → Import Data → CSV,” again confirming that the first line contains variable names. After these steps, the Google Forms responses—now numeric and properly labeled—are ready for statistical operations in SPSS.
Cornell Notes
Google Forms responses exported to Google Sheets often arrive as text (e.g., “strongly agree”), which SPSS will not treat as numeric survey items. The workflow converts each response label to the questionnaire’s numeric coding using “Find and replace,” with “match entire cell contents” to prevent partial replacements (especially for “agree,” which appears inside “strongly agree/disagree”). After mapping categories like gender to numbers and fixing any replacement side effects, the sheet is downloaded as XLS or CSV. Before importing, variable names in the spreadsheet should be cleaned (e.g., remove spaces) and unnecessary columns like timestamps can be deleted. SPSS then imports the file using “read variable names from first row,” producing a usable dataset for analysis.
Why convert Likert responses from text to numbers before importing into SPSS?
What goes wrong when replacing “agree” without special matching rules?
How does the workflow handle gender recoding and a common replacement error?
Why rename variables before importing the XLS into SPSS?
What import settings matter most for XLS and CSV in SPSS?
Review Questions
- When converting Likert items in Google Sheets, what specific option prevents “agree” from being replaced inside “strongly agree/disagree” responses?
- What variable-name issue can occur when importing Excel into SPSS, and how does renaming columns solve it?
- Why might CSV import fail or split text incorrectly, and what precaution is recommended before importing?
Key Points
- 1
Convert Google Forms response text (e.g., “strongly agree”) into the questionnaire’s numeric codes using Google Sheets “Find and replace.”
- 2
Use “match entire cell contents” when replacing overlapping labels like “agree” to avoid corrupting “strongly agree/disagree.”
- 3
Recoding categorical fields (like gender) requires careful matching; fix any accidental value shifts with follow-up replacements.
- 4
Rename spreadsheet columns to SPSS-friendly variable names (avoid spaces) before importing, and delete unnecessary columns such as timestamps.
- 5
Import XLS into SPSS via “File → Import Excel,” ensuring SPSS reads variable names from the first row of the data.
- 6
Import CSV via “File → Import Data → CSV,” confirming the first line contains variable names and watching for commas in text fields.