Sometimes documents are slow because of the size of their Data Set. This could be due to the raw data or the number of new variables created in Displayr. Many times performance can improve by simply saving the raw data file from a new document to a more compressed format (a leaner .sav or .QDat) and updating that file to the Document with your Report.
However, more significant improvements can be seen by using a separate document to perform more of the data preparation outside of the Document you build your Report in, such as:
- Removing unneeded variables and cases.
- Creating constructed variables that do not need to be updated in real-time in the report, such as top 2 box scores, text categorizations, segmentation membership, etc.
- You would not use this for filter variables that depend on user-selections in combo boxes or other controls on the page as they need to be recalculated when the control selection changes. Nor would you use it for constructed Banners as they use a dynamic interface and have extra nets, spans, and other structures integrated into them.
- Stacking data.
- Merging data, such as real-time API data into a historical data file - this is especially handy for tracking studies.
Any categories you have merged, NETs you've created, or categories deleted in tables will not be merged, created, or deleted in the saved file.
Requirements
- A data set loaded into a new Displayr document.
- If setting any data up to export automatically on a schedule, you'll need Server time for private dashboards and API hours (view hours) on your account.
Method
- Decide if you would like to start with the raw data and edit from there, or take the Data Set you've created (with constructed variables) and whittle down your report to what you want to include in the data preparation.
- Starting from the raw data - create a New Document from your Documents page, append something like Data Preparation to the report name, and import your raw data.
- Starting from your current data set - Duplicate the document from your Documents page (via the ellipses button when hovering over the document's name), appending something like Data Preparation to the name.
- In the data preparation document, edit the data to: add in constructed variables that you'd like hardcoded in the report, hide/delete unneeded variables/cases, delete any banner or variables that need to be recalculated in real-time in the report (if working from a duplicated document).
- Save to your cloud drive for the most efficient workflow. Click Publish > Export Data > To Displayr Cloud Drive as Q Data File (.QDat) (the leanest format type as it was created specifically for Q/Displayr) or you can choose To Displayr Cloud Drive as SPSS File (.SAV) if you are using it in the merging or stacking widget.
- On the next screen, select an option below then click OK.
- If you'd like to set up automatic exporting, select Whenever document is published.
- To export manually whenever you want, select Just now.
- Give the exported file a name and click OK.
- [OPTIONAL]: Set up the automatic updating of this data set by scheduling the document to be republished on a schedule (it doesn't matter that it's not been published to a dashboard). See How to Automatically Republish a Dashboard on how to set up the Automatic Updating widget.
- [OPTIONAL]: Perform any merging or stacking that you require using the exported file from your Displayr cloud drive. If desired, use the AUTOMATIC UPDATING sections in their object inspectors to set up a schedule for exporting the merged/stacked data to the cloud (note this will happen whether or not the input data sets have been changed).
- In the original document used for reporting, in the Data Sets tree select your data set.
- Press Update in the object inspector to update it with a new version.
- Click Displayr Cloud Drive and select the appropriate file from your cloud drive that you created in your data preparation document.
- Review any warnings or errors that pop up in the document.
- Repeat steps 8-11 for any other data sets you'd like to update with a file from your data preparation document.
Next
Reduce the Number of Unnecessary Variable and Variable Sets