We have been using downloaded DHIS2 data to construct multi-variable regressions to assess the impact of a comprehensive intervention on maternal and neonatal health outcomes. In the past DHIS2 reported when zeros were entered into the database, but since an upgrade all zeros are being presented as non-entry blanks. This is problematic from a data cleaning standpoint to assess what variables were missing and to check them against hard copy facility registers. It also is hugely problematic from a data analysis and epidemiological standpoint as missing information is not the same as a reported non event. Are there ways to assess what variables were entered as zero and turned into blanks by DHIS2?
1 Like
Hi @blairhoo
Welcome to the community! 
There are a couple of things to check here:
- In the System Settings app → Analytics settings tab, make sure “Include zero data values in analytics tables”
- In Maintenance app → Data Element configuration, make sure under the aggregation type “store zero data values” is selected
After selecting these settings, please test again using Browse Chrome as a guest and run the analytics tables export after you perform Maintenance in Data Administration app (clear analytics tables, clear application cache, and reload apps). I hope this helps!
Were both of these settings selected but you’re still facing an issue? Could you run the analytics tables export again and then check the Catalina.out log for errors? Please share the full log (without sensitive info) if there are errors.
Thanks!
1 Like