# Milestones State Percentile Ranks 2019

The view below converts average test scores for each school into a state percentile rank. This view is helpful for comparing performance over time, across grades and subjects1, and to the rest of the state.

The percentiles show a school’s relative position among all Georgia schools. For example, Burgess-Peterson’s 2019 average 3rd grade English score is in the 85th percentile. This means their average scale score is as high or higher than 85% of schools in Georgia. This is an increase from 56 in 2018.

Use the filters to change to a different district or school, or switch to a more traditional measure like percent proficient. Mouse-over a point to see it’s value, or use the filter in the lower right to label all points.

Source data for this post is from GADOE. If you would like to know about similar posts, please subscribe to the blog.

Data Notes

A few changes were made to the percentile rank calculation this year. The changes are also retroactive to previous years.

• Eighth grade algebra and physical science results are included in the eighth grade math percentile calculation. They are not included in the EOC (high school) percentile calculations. We use the CCRPI content mastery formula2 for eighth grade math and science percentile rank, while we use average scale score for all other grades and subjecs.
• The percentile rank algorithm now follows the Redshift percent_rank calculation, which is similar to Excel’s percentrank.inc function. Values of 0 and 100 are included in this algorithm. We then round results to the nearest whole number, so 0.5% of results will have a percentile rank of 0, and the same for 100, while each number 1-99 will contain 1% of results.

#### Footnotes

1. For example, 2019 state proficiency levels for elementary Milestones range from 31% for fifth grade social studies to 52% for third grade math. A school that finds a 21 point gap in their own performance across grade and subjects might think they’ve found a large gap in instructional quality, when the gap might just be an artifact of varying exam difficulty. Using state percentile ranks helps difference out variation in exam difficulty.
2. The content mastery formula has a 0.99 correlation with average scale score when used to rank schools.