Calculating District CGI for an Overall View of Growth on the MAP Test

NWEA now publishes a Conditional Growth Index score and percentile as part of their suite of reports for districts administering their assessments.  The score allows for comparing student growth across RIT and grade levels so scores are more easily “pooled” together.  The reports do not, however, contain a district level CGI (individual and grade level are standard).

If you would like to calculate the overall CGI for your district, follow the steps below:

1. Download the “Combined Data File” from the Data Export Scheduler in MARC.  If you do not have this option, see your MAP administrator.
2. Sort the resulting .csv file to separate the tests.  Our district administers math and reading, so I end up with two tabs in my spreadsheet, one for each subject.
3. Find the column that indicates the growth period you are observing.  For my most recent data, I was interested in the Fall to Winter growth period so the column I was in search of was titled, “FallToWinterConditionalGrowthIndex”
4. Apply the following formula to the range of cells under that column heading, “=AVERAGE(*:*)”  Mine appeared as follows: =AVERAGE(V2:V1809)
5. Apply the following formula to the cell cell where the average CGI is displayed, “=NORM.DIST(V1810,0,1,TRUE)” V1810 is where the average CGI was displayed, 0 is the mean, 1 is the standard deviation, “True” ensures the formula returns the cumulative distribution function.
6. The above formulas result in the ability to see the percentile of the district’s combined CGI according to normal distribution.

Why might this be useful?  It may offer a different picture of growth health for a district rather than simply adding up “yeses and nos” on an Achievement Status and Growth Report.  It may.  The reason being that in a traditional counting of yeses and nos method, a district may not get the complete picture of all growth.  For instance if we were to average all of the yeses as “1s” and nos as “0s”, then there is no credit for “close” or “knocking it out of the park”.  In looking at CGI, some of your students may have experienced 4 standard deviations above typical growth and there “yes” now counts as 4 instead of 1.  Similarly, another student may have missed their growth target by 1 RIT point and now their “no” may only count for a “-0.01”.  The result?  What might be a more balanced look at overall growth in a school district.

The image below demonstrates our district’s most recent growth results as calculated by the formulas (and others) described above. Hope it helps.