One of the ways that we create a more fair rating system is by accounting for the LEP/CWD students at a school. These are the English Language Learners and Children with Disabilities. On average, these populations of students don't test as well as students that are native English speakers and typically functioning. As a result, schools that have higher than average concentrations of these students are unfairly given low marks, while schools with less than average populations of these students are elevated, not because the school is better, but simply because they have low populations of these students.
The values in this table are derived from the raw data provided by the US Department of Education. When there is a range of values, we average the range. When there is a "LT10" or "LE10" or "GT50" we drop the letters and use the number. Importantly, this is where we calculate the overall adjusted test scores for each school. We do this by subtracting the # LEP and # CWD columns from the # Students column to get the New Total. Then we calculate the total number of students deemed proficient (Grade All / 100 * # Students). Then we calculate the number of students deemed proficient for both CWD and LEP students (LEP All / 100 * #LEP and CWD All / 100 * #CWD) and subtract these values from the total number of students deemed proficent. This number is then divided by the New Total to get our Adjusted Score. Our Adjusted Score is the performance of kids who are NOT classified as LEP/CWD.
These numbers aren't perfect because there are some assumptions, namely we assume the highest possible values when a column is "LT10" for example, we take the average of a range of values, and we assume there is no double counting, meaning there are no students that are both LEP and CWD. We've looked at the extremes in terms of these assumptions, and one thing is clear, making these assumptions paints a more accurate picture than doing nothing at all, which is exactly what the other guys are doing.