Finding population at a percentage of area median income

[stroskey]

I am trying to find how many people live at a certain % of the area median income. I can find the area median income and the population in each income cohort from the census but I can't find how to get specific numbers. For example, if the AMI is $62,000 but the income cohort is from $60,000 -$69,999 how to I get more specific than that? Does anyone have some suggestions? Thanks.

[wahday]

HUD is the one setting these AMI thresholds, so you may try poking around on their site or call your local PJ (Particpating Jurisdiction) office and ask. The reality, though, is I am not sure if they can get that specific, especially in an economy changing as fast as it currently is. It is also complicated by household size, making estimates of actual number of people a little challenging - that is, AMI for a household of 6 is different than for a household of 4. This is total number of people live in the home, regardless of whether they are of working age or not.

You can assume that half the people in the defined "area" are below and half above the AMI threshold (median being the absolute middle number, not the average). However, when you start talking about percentages of AMI, that won't apply, so I don't know what to say about that beyond contacting your local HUD office. But it gives you some benchmark and perhaps they can narrow it down further.

Good luckl.

[ColoGI]

HUD is the one setting these AMI thresholds, so you may try poking around on their site or call your local PJ (Particpating Jurisdiction) office and ask. The reality, though, is I am not sure if they can get that specific, especially in an economy changing as fast as it currently is. It is also complicated by household size, making estimates of actual number of people a little challenging - that is, AMI for a household of 6 is different than for a household of 4. This is total number of people live in the home, regardless of whether they are of working age or not.

Good luck.

That's a cool way to think about it. For a snapshot in time, you need the distribution in that cohort, but you also may want average to compare the skewness of the median average to tell whether there are, say, 147 or 129 out of, say 22,944 there. Skewness is important because the 1% may be skewing the median income (or conversely the bottom 80% may be).

.02

[OfficialPlanner]

While not exactly where you're looking for, these two links may be of help:

30%, 50%, 60%, 80% of AMI

http://www.hud.gov/offices/cpd/affor...s/income/2011/

120% of AMI

http://www.huduser.org/portal/datase...its_50_120.xls

[estromberg]

That's a cool way to think about it. For a snapshot in time, you need the distribution in that cohort, but you also may want average to compare the skewness of the median average to tell whether there are, say, 147 or 129 out of, say 22,944 there. Skewness is important because the 1% may be skewing the median income (or conversely the bottom 80% may be).

.02

Isn't skewness more of an issue with mean than median?

[Brocktoon]

I think I can use ERSI BAO to find that for you. I am off today but PM me next week with the area and I will run it through. The data will probably be from the ACS 2005-2009 survey.

[ColoGI]

Isn't skewness more of an issue with mean than median?

It may - depends on the data. I'm interested in the outcome of the interesting question. One can guess at all kinds of reasons for the question!

[bjamarsh]

I am looking for the same information.