=== modified file 'resources/zm_hmis/readme.txt' --- resources/zm_hmis/readme.txt 2009-11-17 08:53:50 +0000 +++ resources/zm_hmis/readme.txt 2009-11-17 09:06:16 +0000 @@ -2,16 +2,21 @@ characterizes the total number of data elements submitted by a facility, compared to the number expected in each data set. One would not expect that every facility will submit data on every single data element, as there are variations from facility to -facility in terms of what services are available. Additionally, for -some disease specific data elements, the facility may not report +facility in terms of what services are available. + These scripts have been developed in Zambia in order to calculate +an effective facility data coverage rate, that is independent of whether the facility has marked the data set as complete or not. +The reports and scripts make several assumptions that are fairly specific to the Zambian HMIS situtation. + +For some disease specific data elements, the facility may not report anything for a given month, if they had no occurrence of this syndrome. The hypothesis is that most facilities should be rather constant in terms of the percentage of returned data. Some months it may be higher, some lower, but in general, the percentage of submitted data elements per facility should remain relatively constant. Of course the easier situation is when facilities report nothing, which -of course will be easier to evaluate. +of course will be easier to evaluate. If the facility has "zero" reported data elements for a given month, this would indicate that no +data exists in the database for the given time period. -These SQL scripts take a slightly different view on the "data completeness" functionality of DHIS2 itself, which requires the user to mark the dataset as complete. Several BIRT reports have been included along with the SQL script. They will require modification in each country, specifially on the JDBC connection string and username/authentication. +These SQL scripts take a slightly different view on the "data completeness" functionality of DHIS2 itself, which requires the user to mark the dataset as complete. Several BIRT reports have been included along with the SQL script. They will require modification in each country, specifially on the JDBC connection string and username/authentication. Additional modifications to the SQL scripts will also be required depending on the configuration of the country. The script has a few hard-coded assumptions, namely that the primary level where data is entered is at the organizational unit level four (4) and that that the data sets have a periodicity of months. Obviously, these assumptions may not be valid for other countries. For further information on these reports and SQL contact Jason Pickering .