This disparity in school populations was highlighted last year in Indiana when the nation’s broadest voucher law took effect. The first-year results were reported by Niki Kelly in a July 2012 article in the Fort Wayne Journal Gazette.
From the Ledger, by William Hahn:
Not all school populations are created equal. Unfortunately, statewide testing programs, such as the Florida Comprehensive Assessment Test, implicitly assume that they are.
The article pointed out that private schools with a large number of incoming public school (voucher) students saw significant drops in their Indiana Statewide Testing for Educational Progress-Plus (ISTEP+) scores (the Indiana equivalent of the FCAT).
For example, Ambassador Christian Academy in Gary, Ind., accepted 110 voucher students (about 50 percent of its prior-year enrollment) and saw its ISTEP+ passing rate drop from 78 percent to 57 percent. Similarly, Columbus Christian School (Columbus, Ind.) accepted 29 voucher students (about 22 percent of its prior year enrollment) and saw its passing rate drop from 89 percent to 71 percent.
This impact on test scores provides evidence that education does not have a teacher problem. Teachers at the Ambassador and Columbus Christian schools had not changed significantly from one year to the next. What changed was the nature of the student population they were teaching.
While specific reasons for the dramatic declines were not discussed in the Journal Gazette article, a reasonable assumption is that study habits, preparation levels, learning motivation and meaningful parental involvement are factors that separate high-performing and low-performing student populations.
But these factors are never mentioned in articles about student learning and school performance. Such articles wrongly blame teachers.
So, if teachers are not the problem, what is?
To explore this question, I collected data on the percentage of students economically disadvantaged as well as the FCAT math and reading scores for students in the 32 Polk County middle schools.
This data was reported on each school’s website during the fall 2012 term.
I then performed a Pearson product-moment correlation test on these data.
This statistical test measures the nature of a relationship between two sets of data (variables), in this case percentage of disadvantaged students and FCAT reading and math scores.
The test calculates a correlation coefficient (the degree of linear relationship) which falls between +1.00 and -1.00. A positive correlation coefficient indicates that as one variable increases, so does the other. A negative correlation coefficient means that as one variable goes up, the other goes down.
The results show a high negative average correlation coefficient of -0.876 (the range of coefficients was -0.834 to -0.929) between the percentage of economically disadvantaged students and FCAT math and reading scores for grades 6, 7, and 8.
This means that a school with a high number of economically disadvantaged students tends to score low on the FCAT and vice versa.
Remember, a correlation of -1.000 means that there is perfect negative correlation. So, at -0.876, this relationship is close to perfect.
Parsing the FACT data provides a way to see the relationship between economically disadvantaged student populations and FCAT score irrespective of statistical testing.
The top three schools with economically disadvantaged students are Crystal Lake Middle (88 percent), Lake Alfred-Addair Middle (87.7 percent) and Westwood Middle (87.3 percent). They have an average FCAT score of 228.9, which is 10.8 points below the county’s average FCAT score of 239.7.
On the other hand, the three schools with the lowest percentage of economically disadvantaged students are Lakeland Montessori Middle (19.2 percent), Lawton Chiles Middle (24.7 percent) and McKeel Academy (28.2 percent). They have an average FCAT score of 261.5, which is 21.8 points above the countywide average of 239.7.
Clearly, the nature of a student population impacts FCAT scores.
Based on Indiana’s experience, and considering the correlation analysis set forth above, the following hypothesis is suggested: If the teaching staffs at Lakeland Montessori, Lawton Chiles and McKeel were traded for those at Crystal Lake, Lake Alfred-Addair and Westwood, there would be no significant change in FCAT scores at any of these schools.
These data do not suggest that economically disadvantaged students want to do poorly.
I am confident that most have the ability to achieve grade-level learning.
They just do not have support systems outside of school that help them develop learning habits that promote academic success.
This is a root cause of the FCAT-performance problem. Thus, it is where attention is necessary if student learning and FCAT scores are to rise.
It is time to stop blaming teachers for poor FCAT scores and other ills of the public school system.
[ William Hahn has a doctor of business administration degree, with concentrations in both management and accounting, from Nova Southeastern University, and is a certified public accountant with 24 years of accounting and banking experience. He is a professor of accounting at Southeastern University, Lakeland ].