Visual Software Technical Library
Data Quality and Education Funding
March 23, 2018

Data Quality and Education Funding

Schools and school dis­tricts get most or all of their fund­ing from their par­ent orga­ni­za­tions. Depend­ing on their loca­tion, this may be a school dis­trict, a state or some oth­er orga­ni­za­tion. Regard­less of where the fund­ing orig­i­nates, the source of that fund­ing requires account­abil­i­ty in the form of some kind of report­ing. (Most com­mon­ly this is in the form of dig­i­tal reporting.)

Recent­ly, we took a sur­vey of sev­er­al US school dis­tricts to find out how much they spent gath­er­ing and prepar­ing this infor­ma­tion to send to their respec­tive states as required to receive state fund­ing. The answers were sur­pris­ing­ly alike for places that hadn’t used our soft­ware or methodologies.

The Cost

The Cost of Hav­ing Errors in Sub­mit­ted Data

The Oppor­tu­ni­ty for Improvement

In the course of a giv­en year, find the month­ly pay­roll cost for all non-teach­ing staff in the dis­trict, includ­ing every­one (admin­is­tra­tors, bus dri­vers, cafe­te­ria work­ers, librar­i­ans, teach­ing aids – all non-teach­ing staff).

, and it is typ­i­cal­ly a very large num­ber for some­thing unseen which makes it a ripe tar­get for improvement.

If a stu­dent record is miss­ing or if the data for a student’s record does not meet the state’s data qual­i­ty require­ments, the dis­trict may lose fund­ing for that stu­dent. This may cost the dis­trict up to $10,000 or more annu­al­ly for each stu­dent record with errors in it. The prob­lem encoun­tered by most school dis­tricts is that, in many states, the data that needs to be sub­mit­ted comes from sev­er­al dif­fer­ent systems.

When a sub­mis­sion to the state is need­ed, the data from these sys­tems is col­lect­ed and matched, and final­ly, CSV (Com­ma-Sep­a­rat­ed Val­ues) files are cre­at­ed. These files are then sub­mit­ted as a “test run” to the state after which the state noti­fies the dis­trict about errors. The dis­trict then cor­rects the errors in the CSV files. Unfor­tu­nate­ly, these files are dif­fi­cult to edit and the process is very time-con­sum­ing, so this is where a mas­sive amount of work occurs. 

Fur­ther­more, the process is so time-con­sum­ing that nobody ever cor­rects the errors back in the source sys­tems, so the next time the data is need­ed for a col­lec­tion, those errors will still be there.

By using our Verac­i­ty appli­ca­tion, we bring the state’s data stan­dards and their appli­ca­tion to cur­rent data to the desk­top of those enter­ing the data. As the data is being entered, Verac­i­ty checks against the data qual­i­ty stan­dards. Any errors (and warn­ings) will be report­ed back to the per­son who entered the data through Veracity’s user inter­face, allow­ing them to fix the error while the source doc­u­ments are still read­i­ly avail­able, not sev­er­al months lat­er when the data is being sub­mit­ted to the state.

This allows errors to be fixed once and fixed in the source sys­tem – not in the CSV file. Because the source doc­u­ments are still avail­able, this cor­rec­tion takes less than 3 min­utes. This is sig­nif­i­cant­ly less than the 10 min­utes or more on aver­age that is need­ed once the source doc­u­men­ta­tion has been filed away.

Typ­i­cal­ly this 10 to 3 minute error cor­rec­tion reduc­tion can be seen in cost sav­ings even dur­ing the first report­ing cycle. How­ev­er, the large cost sav­ings will become more obvi­ous in the sec­ond report­ing cycle and beyond when most of the errors that used to be cor­rect­ed every cycle no longer show up in the gath­ered data.

Visual Software, Inc.
en English
af Afrikaansar Arabicen Englishfr Frenchde Germanht Haitian Creolehi Hindiit Italianku Kurdish (Kurmanji)ps Pashtost Sesothoes Spanishsw Swahilizu Zulu