My Two Census

Formerly the non-partisan watchdog of the 2010 US Census, and currently an opinion blog that covers all things political, media, foreign policy, globalization, and culture…but sometimes returning to its census/demographics roots.

Problems with the homeless census

Before you criticize this post as coming from a partisan media outlet,, read its claims over for legitimacy, as it seems to be legitimate:

“”We identified concerns with … inconsistent handling of individuals who either (1) stated that they had already been counted, or (2) stated that they had an address,” the IG reported. “We observed 83 enumerations — at shelters, soup kitchens, food vans and TNSOL sites — carried out by 13 local offices. In over half of our observations, enumerators were inconsistent in deciding whether or not to recount individuals who stated that they had already been counted. We also identified inconsistent practices when respondents indicated that they had an actual residential address. In particular, some of these individuals were counted during SBE, while other individuals were told that they could not be counted because they were not homeless. The enumerators’ natural inclination to avoid duplication often contradicted the procedures in the Census GQE manual.”"

Click HERE to read the full article about potential double-counting in the homeless census.

Tags: , , , , , , , , , ,

11 Responses to “Problems with the homeless census”

  1. anonymous Says:

    Thank you, My Two Census for bringing the truth to light. Yes, there was over-counting and under-counting of the U.S. homeless/transient populations.

  2. Al Forteni Says:

    The LARCC told us they didn’t care if we overcounted the homeless and transient populations because it would make up for those who might have been missed. They are fully aware of this problem and encourage the overcount.

  3. Yet Another Enumerator Says:

    This whole issue is, to put it politely, utter bullshit.

    Before working NRFU, I worked GQE/SBE, including going out on TNSOL night. I may not be an expert (I worked as a lowly enumerator), but I do know enough about how the operation was conducted to say that this is simply more ado about nothing.

    First of all, the Group Quarters Enumeration Enumerator’s Manual (D-569.12) says absoltely nothing about respondents who claim that they’ve already been counted. This issue, however, did come up during our training, through the “verbatim” instructions we received. We were told, basically, to politely ignore this claim on the part of the people we were interviewing.

    There are good reasons for doing so; GQE/SBE, unlike the “regular” part of the census, is not so much an exact head count as it is a statistical look at populations that would otherwise be undercounted or not counted at all, particularly homeless people. While it is, of course, possible that any number of the people we enumerated at, say, the soup kitchen we went to, had been counted in the “regular” census (that is, they mailed in their form), the statistical possibility of an overcount is probably more than balanced by the even greater possibility of this population being undercounted. Presumably, all this is taken into account when the final numbers are crunched by the Census Bureau.

    By the way, here’s what the manual mentioned above has to say about enumerating people for TNSOL, in the case where a person cannot be identified or interviewed:

    IMPORTANT NOTE: If for some reason, you are not able to identify a person because they are sleeping or completely covered from head to toe, ask the GQ Contact person or gatekeeper. If there is no GQ Contact person or gatekeeper, you must go ahead and enumerate the person on an ICR [individual census report] through observation. You should write their name as Person 1, Person 2, and so on.

    The same applies to other types of SBE enumeration, where the enumerator can “enumerate by observation” in cases where people refuse an interview.

  4. Amanda Says:

    I worked GQ as an enumerator before working NRFU as a Crew Leader. During our training in GQ, it was told to us “We’re not concerned with over counting as much as we are with under counting these folks.” This was said over and over again when questions would come up in training and in the field.

    One of the soup kitchens we went to would not allow us on their property to enumerate. We were told that we could stand on the opposite side of the street and observe/count from a distance – not allowed to actually SPEAK to those who were obtaining services from the soup kitchen. We were told to “guess their sex, age, race, & origin” and to use “Male 1, Female 2, Female 3, Male 4, etc” entries for their names.

    Then when we went to do the TNSOL, one of the enumerators recognized one of the people at this location as being someone they had “observed” previously at the soup kitchen. They asked what to do and were told “count them at the TNSOL & the soup kitchen.”

  5. Yet Another Enumerator Says:

    @Amanda: What you say jibes with what my experience of SBE. We were told not to worry about double-counting, essentially.

    Since you made your statements without any obvious comment, it’s hard to tell what you’re getting at here, but again: the people who designed and administered SBE already took into account the possibility of people being counted more than once. This has to be obvious, since many of the places we enumerated–soup kitchens, mobile food vans and homeless encampments–have the same sets of people. So I really do not think this is something to make a federal case out of, unless, of course, you happen to be a Tea Party/anti-gubmint type …

  6. end the census Says:

    The whole undercounting and overcounting, and the attitude of the LCOM’s and all the way down to FOS and CL, goes absolutely against the Constitutional mandate which is ENUMERATION, NOT ESTIMATION. They’re always estimating so why pretend they’re enumerating. Why even bother with a census and this enormous, unconscionable WASTE of taxpayer money. We need an amendment to allow for estimating, because estimating is what’s really happening anyway.

  7. Yet Another Enumerator Says:

    Sorry, but with all due respect, you don’t know what the hell you’re talking about.

    Part of the decennial census does involve estimation. It’s called statistical analysis; that’s how you count, as near as possible, difficult-to-count populations like homeless people.

    Regarding the Constitution, to hear the Tea Party types tell it, you’d think the Constitution spells out the exact methodology to be used by the Census, and that it is restricted to simply counting heads.

    Nothing could be further from the truth.

    Here is what the Constitution has to say about the census, from Article 1, Section 2. This is the only part that regulates the census (other language has been stricken from the Constitution or modified, most notably the business of counting non-white people as 3/5 of a person):

    The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct.

    Notice the part I italicized: that means that the Census is exactly what Congress says it is. For what Congress says it is, I refer you to Title XIII of the United States Code. (Which, in case you’re wondering, says nothing about only counting heads, or of not using statistical estimation techniques, or any of the other bullshit one hears about the Census from uninformed or misinformed individuals).

    Like they say, you could look it up.

  8. Embarressed Says:

    6 of 1, half a dozen or the other. So they made up for the idiots who refused to spend .42 cents of the governemt’s money to return their forms to be counted.

  9. Crew Leader Says:

    @ Yet Another Enumerator: I know we’ve had our differences but, I do respect your inclination to see through the BS and deal with fact.

    I do volunteer work with homeless people. Last year our county wanted to get an accurate count of the homeless within the county. On one night, they held several events throughout the county that served great food, gave out bus passes, conducted minor medical services, etc. The events were well attended by the homeless and, because the events were held simultaniously, a fairly accurate count was made.

    I know the CB can’t do this but, it worked for the county.

  10. Duck Fan? Says:

    Anyone else think that the homeless should be turned into hotdogs and then shipped to hungry countries in the Third World? That way, everybody wins!

  11. Yet Another Enumerator Says:

    @Crew Leader: It was my understanding that in some places, the Census Bureau did just that, both on TNSOL night and in other operations enumerating homeless or hungry populations (i.e., soup kitchens, shelters, etc.), by handing out goodies to the people being interviewed. We didn’t have any stuff to give them in our LCO, but it seemed like a good idea to all the people I discussed it with. (Sorry, don’t have any more details: does anyone else here know anything about this?)