Thursday, June 09, 2011

Demographic Task Force

I have been participating on the Demographic Task Force. The purpose of the task force is not so much to help the District improve its enrollment projections as it is about helping the District improve how they communicate their enrollment projections. The Task Force has had two meetings and will have two more.

At the first meeting last week we got a quick education on the concepts and methods for projecting school enrollments.

At the second meeting this morning we got a much more specific data about enrollment trends regionally and in Seattle. While five years ago only about 67% of children born in Seattle showed up in our kindergartens five years later, last year that number grew to 80%. This is a HUGE change. Significantly reduced out-migration is the key factor.

We also got an education on how Seattle makes school enrollment projections. I have found Rachel Cassidy and Tracy Libros, the folks in Seattle most responsible for those numbers to have been plainspoken, honest, and candid about their practices. I always appreciate that. Today Ms Cassidy showed us two specific examples, Roxhill, where the projections were very close to the actual enrollment and Garfield, where the projections were pretty far off target.

The assumptions used to develop the Garfield projections proved wrong, as we all know, but the method by which they were made were solid and the way they went wrong was not very predictable - the big problem came from unexpected enrollment by 11th and 12th grade students. Yes, they underestimated the number of 9th grade students, but not nearly to the same degree. After being walked through the process and the numbers I'm a lot more sympathetic about the District's Garfield projection. I probably owe someone an apology.

17 comments:

juicygoofy said...

It seemed to me that the kindergarten projections have been way off recently because a large percentage of students were not born in Seattle. Right off the top of my head, I can think of 5 of my daughter's 1st grade classmates who moved here sometime between birth and enrollment. Just something to think about.

Charlie Mas Fan Club said...

Charlie, you are awesome. Thanks for taking on the responsibility of keeping everyone in the loop, and for facilitating this forum. I really appreciate your work. You're an asset to SPS.

Greg said...

Thanks for doing this, Charlie. Great to see it.

Can you elaborate on the 11th and 12th grade influx at Garfield? Was the increase much larger than the overall shift of kids from private to public schools due to the recession (due to parents not being able to pay the high cost of private school)?

momster said...

Kudos to the district for asking you to be on the committee, kudos to you for accepting - can't think of a more constructive, productive combination - their data and your turbo-brain.

No accident that it's Tracy and Rachel - two more honest dealers would be hard to find.

Thanks also for the comprehensive report-out.

Anonymous said...

Thanks very much for the update. It's good to know you're participating in this group. One thing that has always crossed my mind when I hear all these projections for increased growth is that at the same time, there are lots of stories about how fewer and fewer families with kids live in Seattle. It amuses me that Crosscut currently has two articles - one called "Will the Last Family Leaving Seattle Turn Out the Lights" and the other called "How Seattle Schools Can Solve Its Capacity Problem." I'd love to hear how these two apparent trends (families leaving but school enrollment rising) fit together. Jane

Anonymous said...

"While five years ago only about 67% of children born in Seattle showed up in our kindergartens five years later, last year that number grew to 80%."

Very interesting. And, also something I think we'd anecdotally experienced when we were enrolling youngest and oldest. Oldest seemed to operating in a scheme where everyone in the neighborhood went to different schools, while many more of youngest's peer group was planning on going to the neighborhood school.

We do live in a neighborhood with a well-though of school, but I see people actively pursuing the neighborhood option. Now, I' waiting to see if the schools continue to work for folks as the kids get older -- we're seeing a number of people who change their approaches in 3rd/4th grade.

Anonymous said...

"While five years ago only about 67% of children born in Seattle showed up in our kindergartens five years later, last year that number grew to 80%."

Very interesting. And, also something I think we'd anecdotally experienced when we were enrolling youngest and oldest. Oldest seemed to operating in a scheme where everyone in the neighborhood went to different schools, while many more of youngest's peer group was planning on going to the neighborhood school.

We do live in a neighborhood with a well-though of school, but I see people actively pursuing the neighborhood option. Now, I' waiting to see if the schools continue to work for folks as the kids get older -- we're seeing a number of people who change their approaches in 3rd/4th grade.

(oops, reposting with a moniker)

(zb)

Anonymous said...

Yes, can you walk us through how the Garfield predictions went wrong? We have bandied a lot of theories about, including ones that broach on conspiracy. So, if the predictions actually make sense now to you, and there's an explanation of how they went wrong, perhaps we can all learn something about the process (and, perhaps, even learn something about how it can go wrong).

(zb)

Dorothy Neville said...

I would also like more data on the Garfield thing. One friend with a senior who took an AP history course this year told me last Fall that because there were so many unexpected Freshmen, the AP history teacher was moved to 9th grade history and therefore the AP kids had a sub for a month. Now she might have been in error, it is possible she thought it was extra freshmen but it was really extra upperclassmen. But still, the difference in her statement and yours is confusing.

I am not sure this could have been completely unpredictable, given the size of the boundary and the historic distance tie breaker though. Also, transferring students could not get into Garfield, but new to the district students at those grades could be guaranteed a seat, yes? Also, the wealth of AP classes at GHS are accessible by upperclassmen, so it could be a draw, even for those moving from private for just a year or two.

I would be most interested in finding out more.

Fiona @ SP said...

Charlie, thanks for the accurate account of the meeting and characterization of Rachel. I was impressed with her solid work, once I was able to hear her describe the process in depth. Perhaps in the future, when she looks at the projections and makes adjustments, she'll ask for neighborhood input as well.

Charlie Mas said...

The District uses, primarily, a cohort survivorship method for predicting enrollment. For example, they have data on how many students have been enrolled in the 1st grade at a school historically and how many have been enrolled in the 2nd grade. Then they compare each year's 2nd grade with the previous year's 1st grade and get a ratio. A ratio greater than 1 indicates growth (net in migration) while a ratio less than 1 indicates net out migration.

That's the science.

These ratios are typically pretty stable, although they can have trends and bumps. Deciding whether a change is a trend or a bump is the art of it.

Over the high school years, enrollment decays. Students drop out and students get re-assigned to service schools so there's a net out-migration.

The number of 11th graders choosing Garfield in 2010 was actually higher than the number of 10 graders there the year before. It was unprecedented. The school got about 21 more 11th grade students than they were expecting. Even stranger, there was a net in migration of 12th grade students. The school got about 42 more seniors than the model would predict.

There were also about 40 more freshmen than their model predicted. That said, they were predicting the greatest participation rate for Garfield that they had seen in the past ten years (the lookback period). Without the distortion of hindsight, it would be difficult for Ms Cassidy to predict the unprecedented. She did predict the most ever. It was still low.

Just the same, there were 40 more freshmen than expected, but there were 60 more upper-classmen than expected. That's what blew the model.

Greg said...

Thanks, Charlie. Basically sounds like the district failed to account in their model for people not being able to afford private school during this recession and returning to public school. A simple survivorship model would not capture that.

Not quite sure I agree that they should not have anticipated that and planned for this recession and people returning from private school. But thanks for elaborating on the details of why what happened happened.

Jan said...

Well. Hum. Nuts. You know, Charlie, I have been a pretty scathing critic of the GHS boundaries. I do owe someone an apology. I am assuming it should be directed to Rachel and Tracy (and I am assuming their emails are on the District website). Anyone else who should be in the "to" line of this particular email?

kellie said...

I also attended the meeting.

One of the complicating factors mentioned was the enrollment in private schools had also increased. So the Garfield enrollment numbers could not be directly attributed to the return of students from private school.

I suspect that many of the "surprise" new upper grade students were returning from other public districts (ie Mercer Island or Bellevue) and genuine new migration.

Typically families that are relocating to Seattle for employment purposes tend to avoid Seattle because of the open enrollment rules. Under the old plan, families that relocated to Seattle over the summer could typically only enroll their students into last choice schools. Many families that arrive in Seattle over the summer were shocked to learn that their student was assigned to Rainer Beach without any transportation.

Recruiters reinforce this and will direct new hires with families to move over the summer into the surrounding districts where new students get a guaranteed seat.

I suspect that many families new to Seattle moved to the Garfield area based on the school reputation. So I actually disagree with Rachel's analysis.

She did a brilliant job of projecting the reasonably expected and her numbers were amazing accurate for what she was doing. She very accurately estimated the behavior of folks that were already in the system. However, there was no true analysis for folks not in the system.

I went to dozens of meetings about the NSAP. At every single meeting, people would ask if anyone had considered the impact of families moving into the attendance areas for popular schools.

Tracy Libros, who is impeccably honest, clearly stated at public meetings and at board meetings, that the boundaries were drawn using standard housing density models (used by all districts) and historical capture rates. Multiply the number of blocks times the capture rate and you can draw a boundary that matches the capacity of the building.

This worried all the parents on the ground that know that families will move for a good school. Families moved for the good school.

nacmom said...

I too was at this meeting & pointed out that GHS is a high profile example of what Kellie points out,but that it is happening, on a smaller scale, at dozens of other schools.

The million dollar question us whether the district can and will make changes year to year based on recent data? Ie at schools that blew past their 5 yr projections in year one, will they assume continuing increased growth ( they should) and make systemic changes ( boundaries, etc) to head off compounding the existing problem, or just act surprised every year that the prev yr was not an anomaly.

There is an impressive amount of knowledge and willingness to provide solutions by the committee members in the room. Will they hear us? Will they acknowledge the immminet crisis?

Will anyone be accountable for planning vs mitigating?

Stay tuned...

nacmom said...

I too was at this meeting & pointed out that GHS is a high profile example of what Kellie points out,but that it is happening, on a smaller scale, at dozens of other schools.

The million dollar question us whether the district can and will make changes year to year based on recent data? Ie at schools that blew past their 5 yr projections in year one, will they assume continuing increased growth ( they should) and make systemic changes ( boundaries, etc) to head off compounding the existing problem, or just act surprised every year that the prev yr was not an anomaly.

There is an impressive amount of knowledge and willingness to provide solutions by the committee members in the room. Will they hear us? Will they acknowledge the immminet crisis?

Will anyone be accountable for planning vs mitigating?

Stay tuned...

kellie said...

nacmom makes a great comment.

At the meeting they outlined that one of the issues with high school is that historically 9th grade is the high water mark for most high schools. In other words, the 12th grade cohort is much smaller than the 9th grade cohort.

All of the models are based on this premise. They are uncertain as to what the new model should be. I think it is much more likely that the new model more closely resembles the model at surrounding high schools.

It is very likely that the enrollment model for high performing high schools looks more like Bellevue High School and Shorecrest. Does anyone know what their high school enrollment patters are?