Exporting data

The main Export Data page is where you define the data format that you want to use for the output data file, and the type of data you want to export to that file.

Export Data uses the project’s DataLocation property in DPM to find the questionnaire definition (.mdd) file and the case data. If DataLocation points to a project folder in the User area and there is no .mdd file there, Export Data will instead look for a .mdd file in the Shared project folder.

If the .mdd file contains more than one version, Export Data uses the superversion; that is, a combination of all versions, which results in data for all questions that appear in any version being exported.

If you are exporting data in Quantum format and Metadata Model to Quantum has previously been run on the case data, Export Data uses the column allocations made using Metadata Model to Quantum rather than making new allocations.

Note If you downloaded the .mdd in order to run , remember to upload it into the shared area before running Export Data.

1 In Export to, choose the data format in which you want to write the exported data. Hierarchical data is supported in the UNICOM Intelligence Data File and UNICOM Intelligence XML Data File formats.

When the MDM document contains unbounded loop questions, and you select UNICOM Intelligence Data File or UNICOM Intelligence XML Data File, the export will use HDATA (otherwise VDATA is used). For more information, see Unbounded loops.

2 If the data format allows more than one file type, use the File type box to select the one you want to use.

3 If the questionnaire definition (.mdd) file contains more than one language, use the Language box to select the language you want to use for question, response, and other texts. The default is the project’s default (base) language.

4 In System variables, choose how you want to deal with system variables. You can export all system variables, no system variables, or only the common system variables (see Exporting only common system variables).

▪In Variable types choose Select below. Export Data displays one checkbox for each variable type below the Variable types box.

6 If you are exporting data in Quantum format, a set of prompts to do with card column options is displayed. Do the following:

▪In Width of serial number, type the number of columns to reserve for the respondent serial number. The default is five columns.

▪In Width of card number, type the number of columns to reserve for the card number. The default is two columns.

▪In Max card length, type the maximum number of columns to write per card. The default is 80.

▪In Full name of serial variable, type the serial variable name. The serial variable stores the respondents’ serial numbers.

You must export to a *.dau file when the data contains unlimited length text questions, otherwise the export will fail. For limited length text questions, you can export to a *.dat or *.dau file.

7 If you are exporting data in delimited text format and you want to export category names rather than values (codes), click Output category name rather than value for delimited text file.

8 Select Generate extra columns for start and finish times in a specific time zone to calculate and then generate two extra data columns that contain the start and finish times for each case. You can select the following time zones:

▪Interview Server time zone - when selected, the Interview Server’s time zone is used as the start and finish time zone values.

▪Respondent time zone - when selected, the respondent’s time zone is used as the start and finish time zone values.

Note When daylight savings time is in effect in the respondent’s time zone, an offset of 60 minutes is applied.

▪Other - when selected, you can select any start/finish time zone from the provided Time zone list. The default time zone is (GMT) Greenwich Mean Time.

The Filter Options section allows you to include only records that relate to a specific filter condition (such as time period or survey status). Export Data provides several preset filters that you can modify to suit your requirements as well as shortcuts to support filtering based on status and date time. You can also create your own filters, or delete preset filters that are no longer needed.

▪Edit presets: Create, change or delete a filter. For more information, see Creating, changing, and deleting filters: Edit Preset Filters dialog box.

▪Filter expression: Display the expression for the selected preset filter. Click Edit Presets to modify the expression. For more information, see Creating, changing, and deleting filters: Edit Preset Filters dialog box.

2 Under Select filtering based on data collection status, choose Respondent data only, Test data only, or All data, and then at least one of the interview result types.

3 Optionally, enable Select cases finished after a certain date/time and/or Select cases finished before a certain date/time.

▪When Select cases finished after a certain date/time is selected, and an date/time is defined, cases that finish after the specified date/time are returned.

▪When Select cases finished before a certain date/time is selected, and an date/time is defined, cases that finish before the specified date/time are returned.

▪When both Select cases finished after a certain date/time and Select cases finished before a certain date/time are selected, cases that fall between the two specified dates/times are returned.

Note The defined filter condition date/time options do not correspond to the selected Respondent time zone. When defining the Select cases finished after a certain date/time and Select cases finished before a certain date/time options, the times are based on Coordinated Universal Time (UTC).

▪Export Data does not normally export dirty data but can be configured to do so. Speak to your administrator if you need to export dirty data.

▪During an export, DMOM removes all fields, reintroduces the required fields, and adds the fields to the Paper routing (only when the fields are appropriate in the Paper routing). This can result in duplicate items being added to the Paper routing. In DMOM, the generated MDD is not useful for surveys. As such, changes to the routing script are irrelevant.

▪During an export, DMOM removes all variables first, and then adds variables that exist in Query statement. This behavior might cause the exported MDD to contain invalid routing items. As such, the exported MDD must only be used for data export and analysis purposes, and should not be used for survey or project design.