School preference and offer odds
![[Post New]](/jforum/templates/default/images/icon_minipost_new.gif){kind=icon}
Anonymous
|
This year the lottery results data has a new field - "applications on results day." I've been thinking about how you might use that data to get a better understanding of school preferences and/or odds of getting an offer. Some thoughts (welcome others!):
For individual schools/grade levels, you can determine the percent of total applicants that were matched at a school they ranked higher on match day. This gives a rough sense of a school's desirability - the lower this number, the higher the school is ranked as a whole on students' lottery lists. You can also determine the percent of total applicants that were matched at the school or a school they ranked higher on match day. If you assume students are generally ranking similar schools and in a similar order of preference (this is a BIG if), it gives a rough sense of the chances of matching at a particular school or a school you ranked higher on match day. The above kind of falls apart for high school because of how application schools manage their matches and waitlists. But at least we finally have a solid number on how many students are applying to the application schools. You still can't really calculate an acceptance rate though, since some students were presumably accepted to multiple application schools but only matched to one. |
Anonymous
| It is super duper interesting. I wouldn't want to use it without a nuanced understanding of how the lottery has different trends in different grade levels (5th and 6th being high-mobility years, 7th and 8th being low-mobility, for example), and how different schools have such different offerings and feeders. But I do enjoy the data! |
Anonymous
|
5th grade lottery likelihood of matching at this school or a higher ranked school on match day (likelihood to a higher ranked):
Latin I - 32% (20%) Latin II - 35% (21%) Capitol Hill Montessori - 50% (36%) BASIS - 50% (21%) ITDS - 57% (34%) BASIS EA - 67% (33%) KIPP KEY - 80% (51%) DC Prep Anacostia - 88% (42%) TR Young - 90% (53%) Latin I EA - 92% (38%) KIPP AIM - 95% (35%) KIPP Honor - 95% (39%) DC Prep Edgewood - 95% (52%) CMI - 96% (60%) Capital City - 98% (28%) Statesman - 100% (35%) Friendship Online - 100% (39%) I filtered out schools with fewer than 10 matches. |
Anonymous
| Do you mean the Tableau waitlist data? Or did I miss a more current / detailed data source? |
Anonymous
Is this subtracting out sibling preference matches? Spots for Equitable access? Etc. |
Anonymous
|
6th grade lottery likelihood of matching at this school or a higher ranked school on match day (likelihood to a higher ranked):
Hardy - 45% (37%) SH - 55% (41%) SWW@FS - 58% (47%) Sojourner Truth - 60% (46%) KIPP KEY - 64% (38%) ITDS - 68% (50%) Capitol Hill Montessori - 71% (59%) Wells - 78% (44%) Jefferson - 83% (62%) CMI - 86% (67%) DC Prep Edgewood - 86% (55%) EH - 90% (45%) I filtered out schools with fewer than 20 matches and didn't list any of the schools with more than 90% match to school or higher ranked school. DCI doesn't come out in above but looks as follows: Chinese YY - 100% (0%) Chinese - 35% (30%) French Stokes - 56% (0%) French Stokes EE - 85% (8%) French - 33% (32%) Spanish LAMB - 100% (0%) Spanish DC Bilingual - 87% (3%) Spanish MV - 100% (4%) Spanish Stokes - 89% (5%) Spanish - 30% (29%) |
Anonymous
It is the Tableau waitlist data. I downloaded it as an .xlsx from here: https://www.myschooldc.org/resources/data Scroll to where it says "Interested in downloading the My School DC Tableau data? Click here." |
Anonymous
No. However the data is there to be able to do this. FWIW none of the schools listed above have an EA preference. Some do have a separate EA lottery. |
Anonymous
|
As noted above, 9th grade is weird. It's hard to get a sense of what's really happening at the application schools and many other schools that are highly sought after offer no/very few seats on match day. It's still somewhat interesting to see though that about half of the students applying to these schools are still getting elsewhere (presumably the application schools), so I'm sharing the top 15 anyway.
9th grade lottery likelihood of matching at this school or a higher ranked school on match day (likelihood to a higher ranked): DCI Spanish - 45% (45%) DCI Chinese - 47% (47%) DCI French - 48% (48%) Latin - 49% (48%) JR - 53% (53%) MacArthur - 54% (54%) Latin EA - 54% (52%) MacArthur EA - 56% (56%) BASIS - 57% (57%) Capital City - 61% (56%) Sojourner Truth - 62% (56%) Coolidge - 64% (64%) BASIS EA - 66% (66%) EL Haynes EA - 69% (66%) EL Haynes - 72% (64%) With respect to the application schools, here's how many total applicants (percent matched): SWW - 1527 (12%) Banneker - 1291 (19%) McKinley - 1271 (19%) Phelps - 461 (28%) Bard Early College - 387 (21%) Coolidge Early College - 219 (9%) Ellington Visual Arts - 152 (10%) Ellington Theater - 118 (6%) Ellington Dance - 106 (19%) Ellington Instrumental - 105 (19%) Ellington Cinematic - 74 (20%) Ellington Vocal - 56 (38%) Ellington Tech - 46 (9%) Ellington Museum - 39 (38%) At the application schools I'm also wondering if the difference between matches and lottery seats indicates how many students would have matched but instead matched at higher ranked school. For instance, Phelps had 273 lottery seats but only 130 matches. Banneker 260 lottery seats but only 245 matches. Are those 15 and 143 deltas just students who would have gotten in but matched elsewhere? |
Anonymous
Believe the odds for Latin and Basis drastically drop once you account for siblings |
Anonymous
Doesn't seem correct.. For instance, Latin 1 5th grade - 75 matches out of 629 applications. Doesn't amount to 32% odds. What math are you doing? |
Anonymous
...and out of those 75, 29 were preference spots (sibling), so only 46 spots remaining to take with 600 applicants. |
Anonymous
I think PP was counting the kids who matched with somewhere they ranked higher in the numerator. |
Anonymous
Not seeing where the data gives you this information... |
Anonymous
For percent of students who matched at a higher ranked school: (total applicants - matches - waitlist)/total applicants For percent of students who matched at this school or a higher ranked school: (total applicants - waitlist)/total applicants The number the way you're calculating it isn't really meaningful here because it also includes people who didn't match because they ranked another school higher. |