A Gift to VDOE From a Critic

by James C. Sherlock

I write about education in Virginia.  

I think VDOE is going down the wrong rabbit holes in some of its programs, but I want to be able to see what the data says.  

There are vast troves of state and federal government education data, but I have been frustrated by the lack of useful visualization tools for assessing the true academic performance issues in our K-12 schools.

So I have built one. Hardly perfect, but it works.  And no, you have not seen this one before from me in a previous post. I had to change and expand it to make it work the way I wanted.

It now does.  I offer it to VDOE for free.

I experimented and found that the spreadsheet has to get pretty deep into the school divisions to be of any real use, but it also needs to be able to offer useful information quickly.

So I designed and compiled the attached spreadsheet as a project to demonstrate how data can be accessed from government sources to help visualize the performance of a school district, its individual schools and student cohorts within each school in comparison to both statewide schools and to other schools within a district.

The demo version showed me things that I would not — could not — have guessed otherwise. Such a tool can dispel commonly held misconceptions about education and allow people to discover  real issues with real data.

I used the Standards of Learning test scores from 2018-19, the last year undisturbed by COVID, as the baseline to create this visualization and demonstrate its potential utility. I used text bolding, color coding and cell border pattern differentiations to make top-level results visually apparent.

I chose Chesapeake City Schools for the demo.

Chesapeake has nearly 41,000 students and is a close match demographically for Virginia schools statewide. In 2020 it matriculated a student population that was 43% white, 32% Black, 11% Hispanic, 8% two or more races and 3% Asian heritage. There are other demographic data but those define the bulk of the students.

I mapped student cohorts in each of 38 Chesapeake school schools to their pass rates in math and reading SOLs in 2018-19 and compared them to the pass rates for identical statewide cohorts on the same tests. The cohorts were:

(Asian students outperformed in Chesapeake schools at the aggregate level but are not listed for individual schools because most schools had cohorts of Asian kids that statistically were too small to warrant listing their SOL pass rates for each school.)

I found it necessary to go to grade level detail to get meaningful results  I used SOL pass rates for 4th (elementary, intermediate and primary schools), 7th (middle schools) and 12th grade (high schools) as reference points. Someone with more time could do it for every grade.

If such spreadsheets are prepared by VDOE for each district in the state, they also can be employed by a wide range of users to assess comparative results and comparative policies and teaching methods.

Once the design is finalized, the production of the spreadsheets can be automated and the results delivered annually by the Department of Education and provided for download to all interested parties through the VDOE statistics web pages.

I recommend that course of action. This demo spreadsheet is my gift to the government. (Unhappily, some might say I am asking what it is worth.)

A few observations from the sample Chesapeake Schools spreadsheet using the 2018-19 school year SOL results:  

Chesapeake runs an extraordinarily successful school system. If I were the Superintendent of Public Instruction, I would visit Chesapeake city schools to see what they are doing.

Chesapeake schools significantly outperformed statewide student SOL performance in every cohort on both math and reading SOLS.

The performance of all Chesapeake cohorts in math SOLs at every level was extraordinary.

Both White and Hispanic students massively outperformed their counterparts statewide in math.

Cohorts of non-Hispanic Black and Disadvantaged students in Chesapeake each outperformed their peers statewide in both math and reading.

Black student pass rates blew away the statewide averages for that cohort. In six schools, Black students outperformed their white schoolmates in math head-to-head. Similarly, in two elementary schools, Black students outperformed their white classmates in 4th grade reading.

Hispanic students performed as well as the white students in math, and nearly as well in reading.

The Disadvantaged cohort greatly outperformed statewide disadvantaged pass rates in math.

The weakest performing cohort was white students in the reading SOLs, but there, too, that cohort exceeded statewide pass rates in most Chesapeake schools and for the division overall.

The following schools outperformed state pass rates in every test of every cohort:

• Butts Road intermediate,
• Cedar Road Middle,
• Deep Creek Central Elementary,
• Deep Creek Elementary,
• Deep Creek High,
• Edwin W. Chittum Elementary,
• Grassfield Elementary,
• Grassfield High,
• Great Bridge Intermediate,
• Greenbrier Middle,
• Hickory High,
• Hickory Middle, and
• Thurgood Marshall Elementary.  Thurgood Marshall performed with exceptional distinction by turning in that performance as a Title 1 school.

On the down side, every school system has work to do, even one as extraordinary as Chesapeake’s. Oscar Smith Middle School significantly underperformed state pass rates in every test in every racial and disadvantaged cohort. Truitt Intermediate underperformed in most. But that was a couple of years ago, and I suspect the work is well underway.

Everyone who looks at this will see something they did not assume to be true. But these facts are just that. Make of them what you will.