Removing deleted respondent data

Snap stores respondent data in a file with the survey name and extension .rdf (respondent data file). Replies for each respondent are held in individual records in that file. When respondent records are deleted, they remain as empty records in the file. Although they do not affect the analyses, it is sometimes useful to remove the empty records completely. This worksheet shows you how to do that.

Background

Data for each respondent is keyed in or scanned in from a paper questionnaire, or collated from a web or other electronic questionnaire and recorded in the rdf file for the survey. Each respondent’s responses are stored in a single record and each is allocated a unique number known as the case number. The case number is simply the record number of the data for that respondent. Since the case number is unique, it can be used as a reference back to the respondent. This is especially useful for surveys which use paper questionnaires – by writing the allocated case number onto each questionnaire it is possible to create an audit trail from data in Snap back to individual, physical questionnaires.

If respondent data is deleted for any reason, the appropriate record is cleared out but the space occupied remains in the file, thus ensuring that the case numbers associated with other respondents do not change. Deleted cases are ignored for analysis but still appear when browsing through the respondent data. It is possible to remove the deleted cases entirely.

Summary of steps

To remove deleted cases entirely you will need to clone the survey and import respondent data from the old copy to the new one.

Step 1: Clone the survey

At the Survey Overview, highlight the appropriate survey and choose Edit | Clone. Give the new version of the survey an appropriate name and click [Ok].

Step 2: Build a database link to the old survey

Click on the Data Entry button in the main toolbar (or View | Data) and select the File | Database links option.Select “Import from Snap Survey” and click [Next >].

Click on the [New] button to start the database link wizard.

Select “Import from Snap Survey” and click [Next >].

Highlight the survey that this one is a clone of and click [Next >].

Select “Append All Cases” in the Import/Export options and click [Next >].

The next screen shows a correspondence between the variables in the original survey and the variables in this survey. Every variable in the current survey will be automatically connected to its counterpart in the previous survey. There is no need for you to do anything other than click [Next >].

At the final page, give the database link a suitable name and label then click [Finish].

Step 3: Use the database link to import respondent data

At the Database link dialog, select the database link you have just composed and click [Run].

The data will be imported from the previous survey but now omitting the deleted cases.

The final report shows the result.

If there is a topic you would like a worksheet on, email to snapideas@snapsurveys.com