Recover Missing Data due to File Overwrites While the Study was Active

Information can be found in various ways when data is missing due to file overwrites while the study was active.

1) The data (tr) should be in the archive folder. There will be a <study> subdirectory and then the TR files created by date.

2) You can also recover the data from the interviewer logs or the htmlresp logs found in the $CFMC/intv_logs directory.

3) The quota file will also have a copy in the archive folder, but in the shape it was in when the study got shut down. Since quotas probably have been updated since then it may not be usable. You can look for LL18 records in the server log file to find all the quota updates that were done. However, the easiest and safest thing to do is probably just run a scan against the full data set (after merging in the data from the archive folder) to get the true counts.

4) The fone file can be recreated from the server log files using the rdlog*.spx and mrgfn*.spx files in the Survent sub-directory.

5) If playback files are turned on, those files can actually be fed directly into Survent to rebuild the data.

You need to have>playbackfiles in your initial file in the control sub-directory. The playback files will be stored in the directory <studyname>.p_ under the interviewers running directory.