In this article
Response data from other systems or from other Forsta surveys can be imported into an Authoring survey. This enables you for example to use Forsta Reportal to report on surveys created using another system. You may also export data and perform modifications on the data outside Authoring, and then re-import the data.
Note that the dialog and options depend on the format of the file that is to be imported.
If the imported file contains errors, you will receive an email containing two files; one file contains the invalid records, and the other contains an explanation of why the records are invalid. If any invalid columns are discovered during the import, these being system columns or columns not found in the meta data, these errors will be logged in the task log.
Important
If a survey data import task is to be recurring, it must be set to recurring before it is executed for the first time (set start time to "Schedule for later execution" and on next page set it to recurring).
Important
If a survey has loops and you wish to import data to the survey then you can only import Separate Files; Single Files cannot be imported. So if you are planning to import data to a survey with loops, then you must choose the Separate File export option. For example if you wish to perform data cleaning, you must export the data to Separate File, clean the data, then import the Separate file.
Key Fields
When you import data, you must provide a key field that is used as a unique identifier. The key field identifies the records or responses. Typical key fields can be a user id such as a member or customer number, or the system-provided "respid" or "responseid". The key field must exist in both the import file(s) and the database.
You can use Single, Numeric and Open Text variables as the key field. Note that the question must be Indexed in the Question Properties, and an Open Text question must have a Field Width specified in order for the Indexed property to be visible.
Data can be "updated" or "appended". If you select “append” the new response data will be added to the already existing data, if any. If there is a match in the key field between data in the import file and data already in the survey's database, that data will not be added (the original data in the database will be kept). If you select “update” the response data that has a matching key field in the survey's database will be updated. Records in the import file that have no matching key field in the database will be added to the database.
The search for matching values in the key field will also involve matching, and possibly also appending, in the respondent list. This means that importing can also be conducted on surveys that either are live or where data collection will continue later, as well as to panels. If there is no match for the key field in the response data, the import program will move on to checking the respondent list. If there is a match in the respondent list, the import program will fetch the respid from the respondent list and append the import data with that respid into the response data. It will use the respid from the respondent list even if something different than respid is used as key field, and the import files have respids and these differ from those in the respondent list. This is due to respid being an auto-generated, unchangeable id in Authoring. If there is no match for the key field neither in the response data nor the respondent list, and “append” is chosen, new records are added to the respondent list. This will auto-generate new respids that will probably be different from those in the import file.