Two examples of analysis of school districts with the software and data on this web page


The intent is to give specific instructions using the options available to arrive at meaningful comparisons of schools within two districts.  The first example, using Dallas ISD, demonstrates some of the factors that may account for widely divergent performance outcomes among high-poverty schools within the same school district.  The second example illustrates how differences in average general fund expenditures per pupil differs a great deal, comparing expenditures between the elementary and middle schools in Fort Bend ISD, dividing them into two groups based on the percentages of economically disadvantaged students in each school.


Example 1.  Dallas ISD.  With your computer logged on to the web page at, perform the following steps:

1.      Click on the [Hlt/Sel] button if it is not already “on” or red

2.      Choose Dallas ISD from the district select list.

3.      Make sure the [Multi] button is colored red, clicking on it if it is not.

4.      Using the mouse to create rectangles, as described in item 1 in the attached Instructions, choose approximately 30 of the highest performing schools (high STAAR1 values) that have at least 70 percent economically disadvantaged (ED) students.

5.      Similarly, select approximately 30 schools with ED exceeding 70 percent, but with the lowest STAAR1 pass rates.

6.      Observing the two rows of data that were written beneath the diagram, you should see the following, when comparing the data for the lower performing schools with that of the higher performing schools: (a) greater average mobility (MOB),  (b) greater percentage of Black students (B), lower percentage  of LEP students, and (d) higher general funds expenditures per pupil (GFEXP_PP).

7.      The overall pattern of the points in the graph are very similar for the other two largest districts, Houston ISD and Fort Worth ISD. That is, schools are fairly closely distributed around the regression line with increasing poverty rates up to about 70 percent, above which there is an increase in variation around the regression line, both in the positive and the negative directions, at higher rates of poverty. The differences in the percentages of LEP students, especially in elementary schools, and differences in mobility rates, appear to explain at least part of the increased variance at the highest rates of poverty. 


Example 2.  Fort Bend ISD

1.      With the [Hlt/Sel] “on” (red), choose Fort Bend ISD from the district selection list.

2.      Beginning at the extreme upper left-hand corner, create a rectangle that encompasses schools with the lowest rates of ED that include approximately 50 percent of all students in the district (see Item 10 in the instructions included in the Help table)..

3.      Enclose the remaining—higher poverty—schools in a second rectangle.

4.      Comparing the two lines of data, for the higher-poverty schools you should see (a) much greater ED, (b) greater mobility (MOB), (c) greater percentages of Black and Hispanic students, (d) greater percentages of LEP students, (e) lower STAAR1 pass rates, and (f) greater general fund expenditures per pupil (GFEXP_PP) of approximately 800 dollars.

5.      If you capture the three lowest performing schools in another rectangle, still with the [Multi] in the red or “on” mode, you should observe that the rate of Black students is 50 percent.

6.      Another rectangle capturing the 10 schools at the extreme low-poverty end will show, surprisingly, that the total percentages for Black, Hispanic, and White students sum to just 49 percent, from which it is inferred that some 50 percent are Asian American.  Fort Bend ISD apparently has a fairly uniform distribution of students among the four racial groups, but they are obviously not uniformly distributed across its elementary and middle schools.