What school waitlists tell us about the demand for public schools

Last week D.C. Public Schools and the D.C. Public Charter School Board released waitlist numbers for the city’s public schools. The waitlists are a result of the first round of the annual public school lottery, which determines who gets seats at the city’s charter schools, at out-of-boundary traditional schools, and in the pre-K grades at traditional schools.

We took a look at the waitlist numbers to see what they say about demand for public schools throughout the city. The waitlists aren’t perfect measures of demand – the biggest complicating factor is that all open charter seats are filled through the lottery, whereas open seats in traditional schools are offered first to students who live in the attendance zone, then to out-of-boundary students through the lottery. Still, we think the numbers paint an interesting story.

Here are our findings, along with some interactive graphs and maps so you can make your own comparisons and conclusions:

Demand for DCPS schools is strong west of Rock Creek Park and in Capitol Hill. Charters are popular east of the Anacostia River and in the north-central part of the city.

In the map below, each bubble represents the size of the waitlist at a school. The largest bubble represents the waitlist of 1,381 students at Two Rivers Public Charter School at Florida Avenue and Fourth Street NE. Schools that have no waitlist and are represented by small dots. (Click here to interact.)

map

The charter schools with the longest waitlists are Two Rivers and Mundo Verde. In DCPS, Oyster-Adams and Brent Elementary have the longest waitlists. Demand varies by grade.

The interactive graph below shows waitlist numbers by grade. By looking at entry grades (like PK3, K, 6, and 9) you can see which early childhood, elementary, middle, and high schools are in demand. (Click here to interact.)

barchart

Schools with similar test scores can have very different waitlist numbers.

The graph below shows that a lot of schools with higher test scores also have high wait list numbers, but there are plenty of exceptions. Some of the waitlist differences are due to some apples-to-oranges comparisons we’re making, since we’re looking at both charters and DCPS schools; mixing early childhood, elementary, middle, and high; and have even thrown in application high schools.

Despite these differences, you can use the graph to find relatively similar schools with similar test scores that have widely different waitlist numbers. Ludlow-Taylor Elementary and Brent Elementary, for instance, are both DCPS schools where 77 percent of students are proficient or advanced in reading. Ludlow-Taylor has a waitlist of 413, while Brent’s is 880. This is consistent with other research that shows parents consider more than just test scores when choosing schools. (Click here to interact.)

scatterplot

What exactly is this data? The waitlist numbers are from D.C. Public Schools (available here) and the D.C. Public Charter School Board (available here). Test score data and geographic data for schools (ward, latitude, and longitude) are from the Friends of Choice in Urban Schools (available here). The reading scores used in the last graph are the scores from the 2014 DC CAS. Note that the number of applications schools received are even larger than the waitlist numbers since students matched with one of their higher-ranked schools are removed from the waitlists of schools they ranked lower.

Correction: In the original version of this post we were missing the waitlist numbers for the Spanish immersion program at Tyler, grades 2-5, and for the dual language programs, grades PK3-1, at the following schools: Cleveland ES, Marie Reed ES, Oyster-Adams, and Powell. This led us to undercount the waitlists at these schools in our original post. We’ve updated the post, map, and graphs to correct for this.

14 thoughts on “What school waitlists tell us about the demand for public schools

  1. I think you have missed Powell ES, for dual language and as a by-right school, in your analysis here. I’m not sure whether other schools aren’t in your analysis.

    http://dcps.dc.gov/DCPS/Files/downloads/Learn-About-Schools/SY15%2016%20Lottery%20Excel%20Totals.xlsx

    If I read it right, the above-mentioned data from DCPS shows that there were 742 applications for SY15-16 and 363 applications for Powell dual language at PK3. It might be possible that seats offered and matched information isn’t in this sheet and may not be on hand for you all.

    Like

  2. Thanks for pointing this out. We can add this to the charts. I took another look at the DCPS spreadsheet – the issue is that they did not provide total waitlist numbers for some grades in the following dual language programs: Cleveland, Marie Reed, Oyster-Adams, Powell, Tyler. But we can sum up the numbers and add them to the charts.

    Like

  3. absolutely, waitlisted, not applicants. Since it looked like there was a data gap, I just mentioned what I could see. The dual language program/two waitlists issue seems to confuse things each year.

    Like

  4. and just to be clear, it seems like that sheet was missing something, e.g., while there were hundreds of applicants for the dual language schools, column T of R1 Applicants waitlisted was empty if I read the sheets right, and we can’t just sub in zero for those cells. If there are, for example, 260 applicants for Marie Reed at PK3, they didn’t waitlist no one and offer 13 or 14 classes-worth of Pre-K seats (though the seats matched column is also empty).

    Like

  5. OK, we’ve figured it out and have updated the post. To find the total waitlist for the dual language programs for grades PK3-1 you need to sum columns BD through BQ. Thanks for pointing out the missing data!

    Like

  6. I wonder how the results — particularly the correlations — look if you scale the waitlist numbers by the overall school/grade enrollments. (Divide waitlist students by student enrollment).

    A waitlist of 7 in a school-grade with an enrollment of 30 seems different than 7 in a school-grade with an enrollment of 100.

    Like

  7. @Rob: Ah, that would have been a good way to normalize the data. You can try it yourself: if you click on the interactive graph and view it in Tableau Public you can download the raw data.

    Like

  8. I noticed that you added up the wait list totals, but neglected to account for the fact that some schools such as Yu Ying and Latin American Montessori Bilingual do not accept students past a certain grade.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s