Open data can help families with confusing school choices

Educating children in DC's publicly-funded schools can feel like a roller coaster ride for parents who have to sort through myriad educational options. DC can ease this task by making more information freely available and encouraging people to develop more useful tools that help parents make sense of it all.

Photo by Lori Greig on Flickr.

DC provides a menu of options from a growing diversity of programming among DC Public Schools and DC Public Charter schools. "School choice" has become a buzzword, but for many parents, choosing a school is bewildering.

Parents ask friends, spend hours researching schools, identify a few best fits (which may or many not be in the neighborhood), and attend open houses, tours, and student shadow days (all the while pulling the child out of the classroom and taking time off work).

Then, parents have to apply to lotteries, trying ignore the feeding frenzy in discussion forums, and waiting to learn if they hit the lottery and got their child into any chosen schools. Plus, since lottery dates are not coordinated, parents have to go through this angst multiple times.

Many families do not have the time to invest in this process. Nor do they all have equal access networks or information. What's more, families' priorities vary according to needs and values, so we can't really expect the city to provide school information in ways that meet everyone's needs.

A few "school locator" applications have taken some initial steps toward helping parents through this process, but they don't answer many practical questions a family might ask. For example, in my own search for schools, some questions I have asked are:

  • Will there be many families from my neighborhood that I may not know today, but can one day befriend to set up carpools (or bike pools) so my kids can take part in after school activities? (Neighborhood-level data about the school populations would help.)
  • Is the program awesome throughout, or just glossy in year 1? (Data around retention of children at the grade-level would help.)
  • Where do children go when they transition to the next stage? (Data around formal feeder patterns or informal destination schools would help.)
  • What strengths do schools have, aside from test scores, such as particularly good music or arts or sports programs? (Data around enrichment would help, including after school programming.)
To meet the flexibility of needs, we don't need merely more canned reports. We need raw data—data about the students and families of students attending DC schools. In theory, DC's Office of the State Superintendent of Education (OSSE) collects all of this data. (I, for example, fill out 8 or more forms for my 2 kids each year).

Even if OSSE had and could release all of this raw data, many families (including my own) would not have the capacity to sift through the information in a way that would enable them to draw meaningful conclusions. To make sense of raw data we would need technology. Fortunately, many coders in DC are willing to volunteer to use their skills for social good. Examples are Code For DC and events like International Open Data Day, where scores of DC programmers and technologists gather for a day to donate their time towards projects for local, national or global causes.

The bad news is that data is difficult to get. Sure, Freedom of Information Act (FOIA) requests are an option, but they are also a barrier. Legal processes can be lengthy and difficult.

Mayor Gray has made a commitment to transparency. I asked him to open up data sets leading up to Open Data Day.

State Superintendent of Education Hosanna Mahaley Jones and Jeff Noel, OSSE's Data Management Director, responded. They couldn't get me all the data I'd requested, but they sent 12 datasets. In the following 24 hours, Open Data Day volunteers were able to create several tools using this data, which I will discuss in an upcoming post.

The DC government has an opportunity to leverage this energy and to benefit from help by the very community it is meant to serve. The key is that this is possible only when raw data is open and available. With officials' help, DC families can one day soon have tools to help them through the difficult process of navigating school choice.

Sandra Moscoso runs the World Bank Finances Program by day and works on community efforts around education, active transportation, and open government by night. Sandra lives in small, quaint, Washington, DC, where she tries to get a little biking in with her husband and two children. 


Add a comment »

How open will charters be to participating in something like this? They seem to jealously guard information. If there were any move to "mandate" them to share this data, I can see cries of "you're taking away our right to be independent."

by Geoffrey Hatchard on Mar 4, 2013 12:52 pm • linkreport

I'm impressed that they were able to send you as much as they did in such a short timespan.

Much of this information is pretty innocuous and already collected by OSSE, it just needs to be organized and presented to the public in raw form. is a good portal that DC has set up, but it could have SO much more information on it!

As I said on your other post, a student-by-student database with year-by-year what schools they have attended is a non-starter. It would be way too easy to take that data and figure out who an individual student is which opens up a whole host of privacy concerns (mostly fear of singling out students who have moved schools a bunch for whatever reason). You may be able to get something like a school-level view of where kids are from (e.g. how many from each neighborhood) and information about where students go from there (what percentage to each destination school).

@Geoffrey Hatchard
Not sure how much ground charters have to stand on in that regard, they already have to report all this information to OSSE and it is available in public reports.

by MLD on Mar 4, 2013 1:12 pm • linkreport

@MLD-- the "student-by-student" level data that you mention is, as you say, not something that would likely be made available as open data, particularly for those privacy reasons.

However, it is something that OSSE has been building over the last 5 years (at least at the beginning of the process, it was referred to as the SLED, for State Longitudinal Education Database, though I'm not sure if the term is still in use).

It seems possible, though I'm not sure how much it would cost, that OSSE could build some sort of front end system that would allow researchers (or advocates, or parents) to make the types of inquiries that would benefit from student-level data without requiring the granular data itself.

Alternatively, there may be someone at OSSE who can respond to those types of queries as they happen, and provide the required data, manually stripping out any potential personally identifying information, which it sounds like may have happened in response to Sandra's request.

by Jacques on Mar 4, 2013 3:53 pm • linkreport

I'd point out that it should be acceptable to allow any legitimate NGO or researcher to sign a non-disclosure agreement and get the student level data without names. It would retain privacy but allow NGOs to evaluate the data themselves. We are currently dependent on what OSSE will allow. This problem is endemic with the new "reform" movement, as reported at the post.

by Mary Melchior on Mar 4, 2013 6:23 pm • linkreport

@Geoffrey Hatchart. I sure hope not. Transparency shouldn't be harmful to a school nor should it take away independence.

by Sandra Moscoso on Mar 4, 2013 8:48 pm • linkreport

@MLD - thanks for commenting on both. You're probably right, and I wouldn't want to compromise anyone's privacy. So perhaps, to meet my request, the focus should be more around percentage of children... However, I do wish someone (any researchers our there?) would take the data collected about children individually as they move around in the system(s) and visualize patterns. You'd definitely see informal feeder patterns that would hopefully help to inform broad decisions like the coming school boundary exercise, the school closures, and provide a reality check around how much families have to (or don't have to) travel to get what they perceive to be a 'good' public school education for their children.

by Sandra Moscoso on Mar 4, 2013 8:57 pm • linkreport

@Jacques - hi! Yes, a tool built by OSSE would rock. In absence of that, how about a tool built by awesome volunteer data scientists (a la @harlanh) and programmers w/ superskills? Not sure I would want to have to ask for data adhoc. Everything collected that is not sensitive should just be published as open data. I'm happy to help myself, as are many others. Having to ask and wait is a barrier and whether or not by design, a deterrent.

@MLD - forgot to give you a shout out for mentioning - where all DC data should live. Would make it soooooo much easier for anyone in search of it.

by Sandra Moscoso on Mar 4, 2013 9:01 pm • linkreport

@Mary Melchoir - Hi! Exactly. I have a parent's perspective, NGOs may want to advocate or design programming, researchers need to fact-check, other DC agencies may need to plan, entrepreneurs may need the raw resource of data - heck, if the fabulous EV Downey is building a business around advising DC families on how to apply for PUBLIC school lotteries, than there is surely a developer out there who can build a successful app. DC has an incredible opportunity to facilitate all of this by simply opening up its data.

by Sandra Moscoso on Mar 4, 2013 9:07 pm • linkreport


I think you could probably get the useful information you're looking for out of those reports if you asked OSSE to produce a summary data file from that individual data for each year.

If you're trying to look for feeder patterns, then I would suggest you ask them to create a summary file with fields like:
Sending School
Receiving School
Number of Students

So then for each year you know how many third graders went from school A to school B. They would just have to define the year period (probably August-July?) and how you define what grade a student is in. Then set a threshold at which students get bundled into an "other" category, maybe 5? So if 4 3rd graders go from school a to school b they would be counted in two records: 3rd grade school a to other and 3rd grade other to school b.

This problem is endemic with the new "reform" movement, as reported at the post.
Often the problem is that stakeholders push as many school resources as possible to "school-based" positions, so central offices are left with bare-bones staff who unfortunately have plenty of things to do that rank higher (and are more useful to their mission) than responding to FOIA requests. And in this case, OSSE was actually extremely helpful in getting this data out.

by MLD on Mar 5, 2013 8:38 am • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

You can use some HTML, like <blockquote>quoting another comment</blockquote>, <i>italics</i>, and <a href="http://url_here">hyperlinks</a>. More here.

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.