"How many households received their first choice?"

"How many households did not receive any of their top 3 choices?"

"How many children in each school have Proximity A/B?"

If you are a parent in the Champaign School District, you may have asked these questions at some point, or heard others ask. The Kindergarten Lottery is indeed quite confusing, and unfortunately the FIC (Family Information Center) is just a tad lacking when it comes to exposing the moving parts that lie underneath the covers. This article intends to prop up the hood, so to speak, so that we can take a look at the engine of the lottery. I will not be going into the history, relevant as it may be, of the Lottery or Controlled Choice.

The games begin as you try to register your child for the first time in the Unit 4 system. If you navigate the FIC website, you may stumble across the "Elementary Enrollment Packet", a form that can be submitted during the "Application Period". You may also happen to notice other pieces of information about "Controlled Choice" and the "School Application Procedure". In the latter, I want to focus on:

Before we jump in a get our feet wet, let me point out one more thing. Unit 4 also holds informational Forums that address the Lottery and Choice process, and one predominate message that is hammered home is "You have more of a chance of getting into a school if it is not overchosen". While this may be true, the parent is left to their own devices to figure out which schools are overchosen. Schools are overchosen? Yes, indeed they are. By definition, some are also underchosen, but we will see that there are schools that tend to be extreme on both ends of the spectrum.

So let us walk down the path of submitting a valid registration form. Pretend that this is a Hollywood movie and we are going to use special effects and everything.

Take a close look at that form: http://www.champaignschools.org/FIC/forms/Elementary%20Registration%20packet.pdf.

The registrant is asked about "normal" logistics like race and ethnicity (confusing in and of itself), address, gender, and a little bit of history, we also see questions about preschool/daycare and who the child lives with at home. Absent from this form but present on a print version I observed are also questions about the educational attainment of the parents/guardians.

The forms are collected and keyed in by FIC staff; things like Race and Ethnicity turn into codes (which are actually listed on the form), as do other items (aside from names and address). The data is saved, and at some point, the entire thing is shipped off to EnrolledU (a Boston educational consulting firm headed up by Dr. Michael Alves). This is where a "computer system to process applications" comes into play.

So here comes my boilerplate disclaimer: I do not work for, nor am I professionally affiliated with, Unit 4 or EnrolledU by any means. The information that follows is purely inductive analysis based on observations of data that has been delivered to me by the School District via a few FOIA requests.

What comes back is a spreadsheet detailing which students were placed in which schools, and the datum related to that assignment. If you want to look at the data and follow along, hop on over to a public database I put up and invite you to peruse: tinyurl.com/cu4lottery

For starters, let us look at the "LOTTERY DETAIL" (lower left-hand corner, the "*" button to the left of "LOTTERY DETAIL"). Alternatively, you can hit that TSV button near the top in the left-hand pane if you want to dump it to an Excel-compatible file. It is a table of 687 rows, one for each student. The columns across the top are quite telling, so I will start there. There are a lot of columns, and I will only touch on ones that I think are relevant.

Skipping the first few, the first I will point out is "sibgroup". This is a weight for the number of siblings.

SES: Either 'L' for "Low Socio-economic status" or 'NL' for "Non-Low"

ASGSCH: Assigned school - this is a numeric code that represents a school

CHASG: If applicable, this number represents which choice the parent received. A majority are '1' for First Choice.

CCRND: This is a random number - it serves as a tie-breaker between two otherwise equally weighted assignments.

CH1APRND: This is the combined weight for the first choice. Essentially, it is CCRND + CHI1PRISES

CH1PRISES: This is the formulaic output of all weights for this choice; CHI1PRIPTS + sum(all other PTS columns)

CHI1PRICODE: This column is the sum of sibling and Proximity weights.

CHI1PRIPTS: This weight, the number of points awarded based on CH1PRICODE, is the most significant factor in school assignment. So let's explain those weights (next three)

CH1SIB: Does this child have a sibling already at this school? If 'Y', this weight is worth 3.

CH1PRXA: Does this child live within the prescribed, magical 1.5 mile "walking distance" of the school? If 'Y', this weight is worth 2.

CH1PRXB: Does this child live outside the 1.5 mile "walking distance"? If 'Y', this weight is worth 1.

Some notes. Proximity is a huge deal in Unit 4, and I could spend a few pages talking just about that alone. But let us simplify it and just say that if you live close to a school, you have a much better chance of getting into it.

So, you might start asking questions at this point. Like, what if I live within 1.5 miles of 5 different schools? Is it possible to not have Proximity B or Proximity A? Is it possible to not get into a school if a sibling is already there? The answers may surprise you. We can look at the data to answer how each question was handled.

For the person that lives near Carrie Busey (within 1.5 miles of many schools), it all depends on which school you actually choose and the order you choose it in. Most likely you will get in your first choice if you choose one that is closest.

For the Proximity question, you may be thinking it is impossible to not have Proximity A or Proximity B - by definition, you are either inside 1.5 miles or outside, right? Well, you would be wrong. Apparently, according to the data, there is a third class. I have no idea what it is called - Proximity C? Two unlucky students have this classification, ID 439 and 654.

It is indeed possible to not get into a school if a sibling is already there - but highly unlikely, and one might say "statistically impossible". In order for that to happen, all Kindergarten applicants would have to have siblings already, and you just happened to get bumped out.

 

What was this about "overchosen schools"? Perhaps it would be easier to couch this in a little bit lengthier of a description, "which schools statistically have more applicants than available seats?". There is a high correlation between how the school is rated via the IIRC (Interactive Illinois Report Card) and the number of applicants over and above the number of available seats. In 2009, Barkstall was the most overchosen (as measured by Total Applicants - Available Seats), followed by Bottenfield. In 2010, these two schools switched positions. If you look at the lottery data for 2010 ("Which Schools are Overchosen"), you will see that Bottenfield, Barkstall, Robeson and Westview had more parents marking down first choices for these schools than available seats. Even South Side, with one open seat, had more requests than available seats. Statistically speaking, these are the "overchosen" schools, and all things being equal, your chances of getting into one of these schools is less than 100%. For "underchosen" schools, your chance of getting in is very close to 100%. At the other end of the spectrum, for the past two years (the only data I have at my disposal), Garden Hills has had many more seats than requests; 77 "regular program" seats, but only 39 total assignments in 2010! This is amazing, in my book. Kenwood, Stratton and Doc Howard also seem to regularly have over 10 seats left over. Clearly, for whatever reason, a trend exists whereby parents are choosing certain schools over others. It will be interesting to see how these stats change in the next few years as Garden Hills, BTW are upgraded and Carrie Busey is relocated.

So, back to the data and number crunching. What are the weights? Mealpts, incomepts, childpts, adultpts, pedpts and prepts. I can make a guess as to exactly how those points are calculated (as with CH1PRIPTS above), but I believe that is not really constructive at this point in time. The main point is that the things you put on the form make a difference. Granted, one could abuse this knowledge and take advantage of the form. I would be dearly heartbroken if you chose to do that. My goal in spelling all this out is to help us, as a community, understand what is going on behind the scenes and help demystify this black box of School Assignment. I want to make it obvious that there is simply math going on - there are no capricious decisions, nobody with a vengeance.

There are, however, assumptions. And I have been trying to talk to Dr. Alves and Unit 4 about these assumptions. For instance, the formulas assume "Parental Educational Attainment" to be an indicator of something, the higher the degree, the worse. "What?!?", you might ask. Or the more number of siblings and folks living in the household the better. Strange, is it not? I cannot put words into Dr. Alves mouth - you will have to read his material and listen to him speak for yourself. My personal impression is that he is trying to give those in tougher home situations a better chance. Frankly, I am cool with that.

That is a tangent, but an important one. The lottery does not stand by itself; there is a bigger picture and it is immensely helpful to understand that bigger picture. I myself have not fully grasped it, but am working on it. *grin*

Before I close, allow me to make examples of two anonymous kindergarten applicants. I am going to choose two students, represented by two different ID numbers, and see how they fare. Please note that the names are completely arbitrary and ficticious.

Student A, let us call her Jill. Her ID is 298 so you can follow along in the spreadsheet. Jill is a non-hispanic, white girl. Both her parents have a Doctorate or PhD, she went to a private preschool, and she lives close to 4 schools (Carrie Busey, Doc Howard, South Side and Westview) with one other sibling. Her parents chose, in this order, South Side, Westview and Doc Howard. They filed their forms and crossed their fingers. Some time late in April, they hear back - their child was placed in Doc Howard. Looking at the Lottery Data, we can see why she was not assigned to South Side or Westview. Her total score for her first choice was 4275533. South Side only has 46 available seats, 15 of which are for LOW SES. Of the remaining 31, 21 seats were filled with students that have siblings, leaving 10. Of those 10, each student had higher weights. In fact, Jill was one of 21 families that chose South Side as a First Choice but did not receive it. Westview has 54 seats in the Regular program (opposed to the ESL program). 24 taken by LOW SES. Actually, by the time Jill was down to her 2nd Choice, her SES did not matter - Westview was simply out of seats by First Choice picks and those with siblings. The ultimate irony is that South Side ended up with 1 empty seat. I will let you ponder that for a little while.

Student B. ID 337, aka Jimmy, is a Hispanic boy who lives with 3 adults and 2 other children, the highest level of schooling is High School. He does not live within 1.5 miles of any school, but has been assigned a Proximity B to Kenwood. His parents chose Garden Hills as his First Choice. Because so few other parents chose Garden Hills, Jimmy was pretty much guaranteed a seat, and he was assigned to Garden Hills.

There are 685 other examples we could use. Really, it all comes down to a mathematical formula that has been weighted with assumptions. I believe the root question that folks ask is "Is it fair?" Some ask if this is fair for their child. Some ask if it is fair for African American/Latino/Caucasian children as a group. Some ask if it is fair for everyone.

My answer: I believe Unit 4 has excellent Elementary Schools, staffed by wonderful, hard working teachers. Personally, I feel that no matter what school a child is assigned to, they will receive a high-quality education.

Still, I would love to see the FIC take steps of their own to disenshroud this thing called "Kindergarten Lottery".

-- 
Charles Schultz is a local parent and has been very interested/concerned about Controlled Choice. Feel free to email him (sacrophyte at gmail dot com)