Reply to Median Family Household Incomes by School

Forum Index » Montgomery County Public Schools (MCPS)

Reply to "Median Family Household Incomes by School"

Subject:

Emoticons





More smilies

Text Color: Font: Close Marks

[quote=Anonymous][quote=Anonymous][quote=Anonymous][quote=Anonymous]There is no way that’s correct [/quote]

What seems off?[/quote]

Way too low. Is this just from all the people who filed in the zone and then it says how many families at that school. I would guess that there are many, many families at Whitman that make in the high 6 figures. There's no way to extrapolate only the student families. [/quote]

Basically, yes. There's no way to limit it to only student families.  This is all families in the service area.  Families include retirees.

I've been working on an updated version that just looks at families with kids under 18.  I have some initial numbers there, but I've been trying to find ways to double-check them, particularly since some of the differences were surprising. Switching to families with kids under 18 required switching to pulling census numbers from tracts rather than data block groups. High school boundaries are big enough that tracts shouldn't be a problem.  When the reported median is under $200k, there isn't much interpolation involved except for assuming a uniform distribution of incomes within each census-reported income bin.  That's certainly not accurate, but isn't going to dramatically affect the data.

It has to model the incomes above $200k. I kept looking for a way to avoid it, even if it meant shifting a mean instead of a median, but the data isn't available at a granular enough level. There seems to be a lot of blocks where aggregate income numbers are withheld, which usually indicates someone very, very rich lives there. So, I'm much less confident in the over $200k medians, but they shouldn't be far off. Churchill's number is surprising.  This suggests families with kids in the Churchill area have slightly lower incomes than families overall.  I'm skeptical.  This might be a byproduct of how this modeling works, but there's no particular reason it should do much worse with Churchill data.

[img]https://i.imgur.com/wWhaFvR.png[/img]    [/quote]

Options

	Disable HTML in this message
	Disable BB Code in this message
	Disable smilies in this message

Review message

Search Recent Topics Hottest Topics