New Test, Same Results: ILEARN Reflects Family Income

Indiana’s new standardized test, ILEARN, may be new and even “computer adaptive,” but it has at least one thing in common with its predecessor ISTEP+. Scores on ILEARN correspond to socioeconomic status. Put simply: The poorer the families served by your school, the poorer your school will perform on the test. Shocking, we know.

Some news reports about the test talk just about the overall low scores. Others go skin deep by comparing the average scores of schools and districts. But scratch the surface, and you’ll find that this test—despite its price tag of $45 million—delivers more of the same. 

For example, take the 20 schools with the highest ILEARN averages, clustered in the upper left of the following graph. Minus two outliers (one of which had only 30 students take the test), the top-performing schools had 0 to 25% of students qualifying for free lunch, and in half of those schools, fewer than 5% qualified for free lunch. The free lunch rate is an indicator of the poverty level of the population served by a school; across Indiana, the average free lunch rate is 40%. Two of these top-scoring schools—both religious schools that received vouchers—had 0% of students qualify for free lunch. 

When you look at schools in Monroe County—public, charter, and private—you will see a textbook-worthy example of statistical correlation. 

And to hint back at our voucher blog post, here are the ILEARN averages of schools in the Archdiocese of Evansville presented against those of the local public school district, Evansville-Vanderburgh. Look at this graph and ask yourself if this merited the headline, “EVSC below state ILEARN average while Warrick, Catholic schools above.”
​

 
We’ve combined the ILEARN data with free lunch data for schools across Indiana here (the source for both is the IDOE; ILEARN was administered to grades 3-8 so only those schools are included).  This is the resulting scatterplot: 

We invite you to explore this data and share what you learn with the editor and education reporter of your local paper—if you are lucky enough to have a local paper—and with your state senator and representative.

Maybe next year, instead of headlines like, “X schools score well, Y schools score badly,” we'll see headlines like “ILEARN scores reflect wealth and poverty—again.” (See, for instance, this coverage in Ohio in 2017.)

Is the important question “How can we improve schools' ILEARN scores?”
 
Or do we need an array of questions? Such as:

• What are legislators doing to reduce poverty in Indiana?
• When will our lawmakers commit to funding adequate numbers of teachers, counselors, school psychologists, social workers, nurses, and librarians?
• Do children in all our schools, no matter their family income, have access to challenging, advanced classes; to art, music, literature, science, and history; and to experiential learning and extracurricular opportunities? Are we providing a rich curriculum for all students?
• Do our communities have the will to take steps toward integrating schools so that children from across the socioeconomic spectrum learn together?
• Will Indiana’s citizens continue to consent to an A-F system that punishes schools that serve families in poverty?

Now those might just lead to an actual accountability system.

–Keri Miksza and Jenny Robinson


P.S. For those who would like the statistical nitty gritty, Pete Miksza has explained the following to us:

Monroe
Pearson correlation coefficient r = -.70. There is a strong negative (e.g., inverse) relationship between free lunch qualification and ILEARN proficiency. Those schools that have the most free  lunch students also have the lowest ILEARN proficiency rates, and vice versa.

On average, schools experience a .70 percentage point drop in proficiency with each percentage point increase in free lunch qualification. Each 5% increase in free lunch qualifications is associated with an approximately 3.5 point decrease in proficiency. Each 10% increase in free lunch qualifications is associated with an approximately 7 point decrease in proficiency. Etc.

Vanderburgh
Pearson correlation coefficient r = -.91. There is an extremely strong negative (e.g., inverse) relationship between free lunch qualification and ILEARN proficiency. Those schools that have the most free lunch students also have the lowest ILEARN proficiency rates, and vice versa.

On average, schools experience a .65 percentage point drop in proficiency with each percentage point increase in free lunch qualification. Each 5% increase in free lunch qualifications is associated with an approximately 3.25 point decrease in proficiency. Each 10% increase in free lunch qualifications is associated with an approximately 6.5 point decrease in proficiency. Etc.

All
The Pearson correlation coefficient r = -.74. There is a strong negative (e.g., inverse) relationship between free lunch qualification and ILEARN proficiency. Those schools that have the most free lunch students also have the lowest ILEARN proficiency rates, and vice versa.

On average, schools experience a .56 percentage point drop in proficiency with each percentage point increase in free lunch qualification. Each 5% increase in free lunch qualifications is associated with an approximately 2.5 point decrease in proficiency. Each 10% increase in free lunch qualifications is associated with an approximately 5 point decrease in proficiency.