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.

MyTwoCensus Investigation: 2010 Census Response Rates Lag Behind Response Rates From 2000

Correction/Update:

1. I may have misheard Dr. Groves at the Wednesday Press conference when I wrote that he said 2010 response rates were as good as they were in 2000.

2. However, this doesn’t take away from the fact that 2010 response rates are significantly WORSE than they were in 2000. My suspicions were also raised today when I learned that the response rate increased by 14% in one day. This means that some 25 million forms were processed in the past 24 hours, which is historically unheard of!

I apologize for any inaccuracies, but I stand behind the data and statistics that I am reporting, and furthermore, other than the one statement above, I stand by the rest of my claims. I was likely confused when I heard Dr. Groves say “We’re off to a pretty good start.”

Though we don’t have the full transcript yet (we will publish it here as soon as we get it), Census Director Robert M. Groves made claims at yesterday’s press conference that mail response rates for the 2010 Census were ahead of/on par with what they were in 2000. These claims are false for the following reasons…

According to Appendix F of this document from the 2000 Census, http://www.census.gov/pred/www/rpts/A.7.a.pdf, the mail return rate was at 42% ten days after the major questionnaire mailing period began on 3/13/2000. But in 2010, ten days after the process started on 3/15, the  participation rate is at only 20%. Here are screenshots from the 2000 report and from 2010Census.gov to check out the data:

Now, look at the mailback rate for 2010 on 3/25 (This year the mailing started on 3/15. In 2000 it started on 3/13.):

*ALSO, PLEASE  KEEP IN MIND THAT THE 2010 CENSUS FORM IS WAY SHORTER/EASIER TO COMPLETE THAN THE ONE FROM 2000!

Tags: , , , , , , , , ,

34 Responses to “MyTwoCensus Investigation: 2010 Census Response Rates Lag Behind Response Rates From 2000”

  1. Rachel Says:

    It appears that you obtained these data from http://www.census.gov/pred/www/rpts/A.7.b.pdf (return rates). The A.7.a link you provide is for response rates.

  2. Rachel Says:

    And as pointed out by the Pew Research Center, the participation rate calculation falls between the response rate and the return rate.

    http://census.pewsocialtrends.org/2010/new-measure-of-participation-in-the-2010-census

  3. Steve Jost Says:

    Stephen, I encourage you to contact us first before you post so we can explain why your comparison is flawed.

    You are linking to a 2003 report on “Return Rates” and pasting a chart from that report. Return rates are calculated long after the Census is over, after we have completed all operations, including door-to-door followup, because we remove all vacant housing from the numerator. We don’t know the size of the vacant housing stock at this point for 2010. And we won’t until we finish going door-to-door this summer to ensure everyone is counted. In every Census the Return Rate is much higher than the Response Rate by definition. Return Rate is a measure of response from all Occupied housing units. Response Rate is a measure that includes all housing units.

    At no point in his press conference did the Director compare the early data for 2010 with 2000. He only said “…we are off to a good start.”

    Finally, it is tricky business comparing 2010 to 2000 for lots of reasons, but some of the big ones include:

    1. 2000 had a Long Form and a Short Form. 2010 is a Short Form only Census.

    2. We had four data capture centers in 2000, we have three this year.

    3. Our Update Leave operation in 2010 is very different in size, timing and scope than in 2000. This is where we hand-deliver forms to be mailed back to almost 10% of the housing units in the country. They are the earliest returns and why you see states such as North Dakota in the early lead on response today.

    There are many more operational differences, and we will attempt to model and adjust for those and begin making our own comparisons between 2010 and 2000 once we get past these very early collection days. It’s too soon to make judgments about who is ahead or behind, but we believe in a week or so these differences will smooth out and we can make fairer comparisons.

    It does not serve our mandate to get everyone included in the census to suggest responses are higher or better than they truly are. Indeed, the whole point of our publishing Participation Rates in almost real time is to motivate those who have not yet mailed back their form.

  4. Steve Jost Says:

    correction, we remove vacants from the denominator, not the numerator, my Bad. This is of course is why the Return Rate is higher than the Response Rate.

  5. Stephen Robert Morse Says:

    Steve – I am going to stand by my claims. If I am wrong, please tell me why I am wrong. This data seems to be the only published data set available, so that is why I am using it. What other data do you suggest I use? Why did Dr. Groves have a press conference yesterday without mentioning this specific data? Why didn’t Dr. Groves mention the 2000 data? Was he not aware of this data set beforehand?

    Again, I am standing by my claims!

    SRM

  6. Steve Jost Says:

    You have the right to stand by your misstatements. You may want to study other publicly available documentation,

    such as:
    http://www.census.gov/quality/S18-0_v1.4_Response_Rate.pdf

    this:
    http://www.census.gov/dmd/www/response/2000response.html

    and particularly this, and particularly page 7:
    http://www.census.gov/pred/www/rpts/TR11.pdf

    This differences have been publicly disclosed, debated, discussed and studied in academic forums and scientific presentations for decades, ever since the advent of the mailed census in 1970. Please call us if you need more documentation.

  7. statistician Says:

    SRM

    If you are standing behind a claim based on comparing two numbers that you have been told aren’t comparable because the denominators are different (response rate vs return rate) then I can’t take you or this website seriously. Trying to create controversy where there is none then sensationalizing it in all caps isn’t journalism.

  8. tech Says:

    “Again, I am standing by my claims!”

    Standing by your claims after it’s been conclusively shown that those claims are wrong because you’re comparing two different data sets as if they were the same… is idiotic.

  9. Stephen Robert Morse Says:

    No, I am standing by my claims because the government is not releasing comparable data sets and I am working with the best data I have available. SRM

  10. Stephen Robert Morse Says:

    Oh, and if you take everything a professional spin doctor takes at face value, you’ve got to be crazy. I’ll quote David Simon, formerly of the Baltimore Sun and creator of HBO’s The Wire:

    “High-end journalism is a profession. It requires daily full-time commitment by trained men and women who return to the same beats day in and day out. Reporting was the hardest and, in some ways, most gratifying job I ever had. I’m offended to think that anyone anywhere believes American monoliths, as insulated, self-preserving and self-justifying as police departments, school systems, legislatures and chief executives, can be held to gathered facts by amateurs presenting the task—pursuing the task without compensation, training or, for that matter, sufficient standing to make public officials even care who it is they’re lying to or who they’re withholding information from.”

    I’m the only one investigating these matters, so be thankful, not whiny!

  11. Billl Johnson Says:

    Government waste: Do we need 3 mailings ofr a single form? onr letter to tell us the form is coming, then the form itself, and a third mailing is a card telling us we should have mailed the form in. No wonder the country is broke

    What office of ACORN did Mr. Obama find this guy Robert M. Groves in?

  12. Joe Crede Says:

    “A lie ain’t a side of a story. It’s just a lie.” – “The Wire”, Clarifications

  13. statistician Says:

    So, your argument is “I will continue to report and stand behind stuff that I know is wrong until someone gives me stuff that is right.” Scary!!

    There is no value in being the only one investigating something if you are committed to doing it poorly.

  14. TR Says:

    @Bill: Because it’s much, much cheaper to mail those reminders to increase mailback response than to send out enumerators to do the questionnaire in person. There have been studies of this.

    There is plenty of Census waste. What should worry people is the Harris contract, which has been a complete disaster on all fronts. All this other stuff is either not wasteful or small potatoes.

  15. Stephen Robert Morse Says:

    Mr. Jost – I checked out those reports that you posted and they have very little to do with the problems or data that I addressed. SRM

  16. My Two Census » Blog Archive » MyTwoCensus has some competition in the blogosphere… Says:

    [...] Bureau Director Robert M. Groves has been blogging more frequently as of late. However, from the syntax of the blog, we suspect that [...]

  17. Concerned Citizen Says:

    Stephen Robert Morse has performed an important public service by calling attention to the Census Bureau’s historical documentation regarding mail response rates and census return rates. Pending additional technical work by the Census Bureau career professionals (as opposed to the partisan spin doctors who director Groves should have the integrity and courage to silence when it comes to technical issues) anyone can read the documentation identified by Mr. Morse. At the Census web site one can read the reports and consider all the careful qualifications set forth by Census Bureau technical experts, and decide whether the mail return rate or a slightly more expansive measure that reflects new operations, the hundreds of millions spent on address canvassing, etc is the best context to assess progress in the 2010 Census. These data will be critical to assessing whatever comes next in the 2010 Census, and hopefully both the public and the oversight in the Congress, GAO, IG etc. are watching carefully.

    While no past measure would provide a perfect forecast of the present, it is very clear that these past data provide an important context for assessing whatever the Census Bureau says about 2010 response. These data should have been mentioned by the Director, and his partisan political staff should not be allowed to discredit the facts or the truth or those who seek to find or discover it.

  18. JAG Says:

    Empress,

    It has been demonstrated here many times that facts and accuracy are not requirements for posting articles on this website. It is much easier to knowingly represent something as truth or call somebody a liar that to conduct simple research to obtain the facts

  19. My Two Census » Blog Archive » Mail Response Rates Jump 14% in 1 day; 18% in 2 days Says:

    [...] Pretty big turnaround in the numbers, eh? Still, they don’t approach the levels of participation at this point in the process ten years ago. [...]

  20. My Two Census » Blog Archive » Correction/Update Says:

    [...] I may have misheard Dr. Groves at the Wednesday Press conference when I wrote that he said 2010 response rates were as good as they were in [...]

  21. Stephen Robert Morse Says:

    And I have now fact-checked Mr. Jost’s claims: Mr. Jost’s statement that return rates are calculated long after the Census is flat out INCORRECT. Return rates were calculated on a daily basis during Census 2000….this is how the Census Bureau monitors performance and decides exactly when to launch NRFU (non-response follow-up).

  22. will Says:

    I read Jost to be saying that “return rates” are not the same as “response rates,” so comparing them is doing apples-to-oranges. That’s also what the http://www.census.gov/pred/www/rpts/A.7.a.pdf report says on page 17, where it compares the two rates (and shows there’s around a ten percent difference between the two). The governmental jargon is confusing, as jargon often is, but I can see what he is saying after working through it.

    The 2010 “20%” graphic is a “response rate” which gets calculated on an ongoing basis as they get forms back, but the numbers from the 2000 report are a “return rate” which is adjusted for things they learn after all the forms are in like vacancy, and because the “return” number has those adjustments blended into it, they’re not comparable.

    That makes sense. There’s been changes in ten years in America, but not so much that the response rate would be less than half what it was in 2000. When we see something like that, we should start wondering if what we’re looking at is apples-to-oranges, and it seems to be so in this case.

    What we need are comparable “response rates” on a day-by-day basis from 2000, before they did the adjustments. Stephen, can you get those from Mr. Jost or one of your contacts?

  23. Anonymous CL Says:

    Since the Return Rate is calculated in part by excluding vacant housing units from the total, it can’t be determined until after Non-Response Follow-Up, where all the homes that haven’t responded yet are visited and the vacant ones are definitively classified as vacant.

    The Response Rate (percent of all forms returned, not excluding ones sent to vacant units) would be calculated on a daily basis to decide when to launch Non-Response Follow-Up and with what levels of staffing.

  24. Anonymous CL Says:

    The key explanation needed to understand the difference between the 2010 numbers on the Take 10 Map page and the numbers from 2000, is right there on the Take 10 Map page: http://2010.census.gov/2010census/take10map/#PRate

    Why Mail “Participation” Rate Is A Fairer Measure than Mail Response Rate

    The Mail Participation Rate is the percentage of forms mailed back by households that received them. The Census Bureau developed this new measure in 2010, in part because of the current economy and higher rates of vacant housing. The rate excludes households whose forms were returned to us by the U.S. Postal Service as “undeliverable,” strongly suggesting the house was vacant. We will still follow up on all these housing units to ensure everyone is counted.

    Mail Participation Rate is a higher number than the Mail Response Rate we have used over the last decade, but it is a better measure of actual participation and therefore an easier goal to achieve when residents mail back their forms. In 2000, the national Mail Response Rate was 67% and the comparable national Mail Participation Rate was 72%. The Mail Response Rate is important to help us plan for the important door-to-door workload that begins in May during which we visit all households that have not returned a census form.

    It’s also explained at Rachel’s link:
    http://census.pewsocialtrends.org/2010/new-measure-of-participation-in-the-2010-census :

    The mail response rate, the mail participation rate and a third measure of response, the “mail return rate,” are calculated for areas where household residents are asked to mail back forms that were mailed to their homes or dropped off by a census worker. These areas include almost all of the nation’s more than 130 million households.

    The mail response rate is an unrefined measure —the percent of forms sent to households in these mailback areas that are returned to the Census Bureau. It is a preliminary measure that Census officials say somewhat understates participation, though, because many forms sent out by the Bureau cannot be mailed back — for example, those sent to vacant housing units and those where census forms could not be delivered, such as non-existent or non-residential addresses. In 2000, the final national mail response rate was 67%. (The initial mail response rate, over the first few weeks, was 65%.)

    The mail participation rate is a refined version of the mail response rate–the percent of forms sent to households in these mailback areas that are returned to the Census Bureau, after removing from the denominator addresses where census forms are determined by the U.S. Postal Service to be “undeliverable as addressed.” Nationally, the final census mail participation rate was 72% in 2000.

    The mail participation rate is intended to exclude vacant and foreclosed homes, which have grown in number as a result of the national economic downturn. The mail participation rate also may provide an improved real-time measure of participation for areas with large numbers of seasonal homes that are unoccupied on Census Day, April 1.

    The mail return rate, the most precise measure of census participation, is the number of households returning a questionnaire from mailback areas mail divided by the number of occupied housing units that received questionnaires in those areas. It cannot be calculated until the end of the census counting process. At that point, officials will use data from census-takers’ follow-up visits and other sources to total the number of occupied home addresses in areas where residents mail back their forms. Once addresses are excluded from the denominator—mainly for being unoccupied, non-residential or non-existent—the rate will rise. In 2000, the mail return rate was 78%.

    ——

    “Mr. Jost – I checked out those reports that you posted and they have very little to do with the problems or data that I addressed. SRM”

    Relevant information from the links provided by Steve Jost, intended to further explain the distinctions between the apples, oranges, and pears you’ve been comparing:

    http://www.census.gov/quality/S18-0_v1.4_Response_Rate.pdf (Response Rate Definitions):
    Looks like a more relevant link (although likely still not of interest to you) would’ve been this report’s Supporting Documents A, “Variables, Rates and Formulae for Calculating Response Rates and Reporting Requirements: Demographic Surveys and Decennial Censuses”: http://www.census.gov/quality/S18-1_v1.2_Requirements_Demo&Decennial.pdf

    http://www.census.gov/dmd/www/response/2000response.html :

    The Final Response Rates should not be confused with mail return rates. Mail return rates exclude vacant housing structures from the calculation to give a true measure of census “returns.”

    http://www.census.gov/pred/www/rpts/TR11.pdf (Response Rates and Behavior Analysis) page 7:

    The mail response rate as of April 18 is a measure that represents the percentage of addresses eligible for Nonresponse Followup that returned questionnaires prior to the designation of the Nonresponse Followup universe. The mail response rate as of December 31 is a measure of respondent participation by mail in Census 2000. The difference between the two rates is that the December 31 rate was not restricted to returns received before the cut for the Nonresponse Followup universe. Several criteria were used to identify addresses for inclusion in the denominator of the mail response rate. First the address had to be on the Census 2000 address frame (Decennial Master Address File) and be eligible to be contacted in the Nonresponse Followup operation. In addition, the rate was restricted to housing units in mailback areas only. Mailback areas included Mailout/Mailback, Update/Leave, and Urban Update/Leave areas. Finally, addresses that were pre-identified as having inadequate address information for mailout were excluded from the rates. For a housing unit to be in the mail response rate numerator, it had to be in the mail response rate denominator, and the Census Bureau received a questionnaire on or before a specific point in time, i.e., April 18 and December 31.

    The mail return rate is a measure of respondent cooperation by mail in Census 2000. Several criteria were used to identify addresses for inclusion in the denominator of the mail return rate. First the address had to be occupied in Census 2000 (on the Hundred Percent Census Edited File) and be in the Census 2000 address frame prior to the start of the Nonresponse Followup operation. In addition, the rate was restricted to housing units in mailback areas only, i.e., Mailout/Mailback, Update/Leave, and Urban Update/Leave. Finally, addresses identified by the United States Postal Service in Mailout/Mailback areas or by Census Bureau staff in Update/Leave and Urban Update/Leave areas as undeliverable were excluded from the rates. For a housing unit to be in the mail return rate numerator, it had to be in the mail return rate denominator and the Census Bureau received a questionnaire on or before a specific point in time, i.e., April 18 and December 31.

    The main difference between the denominators of the mail response and mail return rates is that the mail response rate denominator includes vacant housing units, addresses determined by the United States Postal Service and Census Bureau staff as undeliverable, and addresses on the Census 2000 address frame which were eventually determined not to exist.

    ——

    My summary of the differences in the rate formulas:

    Mail Return Rate = (forms mailed back) / [ (forms mailed out in Mailout-Mailback areas & hand-delivered in Update-Leave areas) - (mailed forms returned as undeliverable) - (unreturned forms later determined by NRFU enumerators to have been mailed/delivered to vacant/nonexistent housing units) ]

    Mail Participation Rate = (forms mailed back) / [ (forms mailed out in Mailout-Mailback areas & hand-delivered in Update-Leave areas) - (mailed forms returned as undeliverable) ]

    Mail Response Rate = (forms mailed back) / (forms mailed out in Mailout-Mailback areas & hand-delivered in Update-Leave areas)

  25. Anonymous CL Says:

    And notice the last paragraph I quoted from Rachel’s PewSocialTrends link:

    The mail return rate, the most precise measure of census participation, is the number of households returning a questionnaire from mailback areas mail divided by the number of occupied housing units that received questionnaires in those areas. It cannot be calculated until the end of the census counting process. At that point, officials will use data from census-takers’ follow-up visits and other sources to total the number of occupied home addresses in areas where residents mail back their forms. Once addresses are excluded from the denominator — mainly for being unoccupied, non-residential or non-existent — the rate will rise.

    That clearly confirms Steve Jost’s original explanation of the different rates and my 11:50pm comment stating the inaccuracy of Stephen’s 3:11am comment which had claimed:
    “And I have now fact-checked Mr. Jost’s claims: Mr. Jost’s statement that return rates are calculated long after the Census is flat out INCORRECT. Return rates were calculated on a daily basis during Census 2000….this is how the Census Bureau monitors performance and decides exactly when to launch NRFU (non-response follow-up).”

  26. Anonymous CL Says:

    The original source of the problem here, at least in part, seems to have been Stephen’s apparent use/reading of the terms ‘response rate’, ‘return rate’, and ‘participation rate’ as if they were interchangeable synonyms. They may be synonymous in typical casual conversations on other topics, but they have specific distinct definitions when used in the context of the Census. Many other people are likely similarly confused by this terminology, but they all ought to get it straight, particularly if they’re going to call it journalism.

    ——

    Stephen wrote: “No, I am standing by my claims because the government is not releasing comparable data sets and I am working with the best data I have available. SRM”

    Yes, as another Pew article says, “The Census Bureau has not released data showing a comparable figure for the mail participation rate at this point in time during the 2000 Census.” And it would be good if they did release comparable figures. They could put the data into that Take 10 Map application so that you could toggle between a 2000 Participation Rate layer, a 2010 Participation Rate layer, and comparison layer (is 2010 ahead or behind 2000), each day.

    But no, since you were working with data that didn’t mean anything in relation to each other (and getting them mixed up), you can’t/shouldn’t just excuse it as ‘that’s the best I could do’ and stand by it. If the best you can come up with on a certain topic is totally inaccurate/false/meaningless, then you shouldn’t say anything on that topic until you do have something with some actual substance. This ‘best’ was useless, not just ‘not quite perfect’.

    Certainly not the “high-end journalism” you claim to want to provide. Particularly since this, and other related confusions you had, apparently led you to originally post this article (now titled “MyTwoCensus Investigation: 2010 Census Response Rates Lag Behind Response Rates From 2000″) with the outrageous title “MyTwoCensus Investigation: Census Director Robert M. Groves Is Full Of Lies”. You shouldn’t make such bold statements/accusations without firm ground to stand on; without firm ground, posts should be written much more tentatitvely, if even written at all.

    The ‘shoot first, miss by a mile, ask questions later (or not at all), and misunderstand half the answers to those (asked or unasked) questions’ approach demonstrated in this post and your comments on it thus far, is not good for establishing/maintaining a reputation as a respectable source of decently-reliable information. You should rethink that approach/attitude unless you want you and your site to end up only appealing to the certain type of right-wing audience who doesn’t care about accuracy as long as they’re told what they want/expect to hear in as outrageous a way as possible, and you don’t care about alienating everyone else.

    I have been reading and commenting on this site since it began over a year ago; if you tallied up all the comments, I have likely made the most comments, spending quite a bit of time/effort on quite a few of them. When the Census does something wrong/wasteful/dumb/incompetent, I inform, confirm, or agree. But when the Census is inaccurately accused of something or misconstrued by some people, I state the correct information or try to help people understand things. And when the Census does something good/reasonable/competent, I admit it, compliment it, or at least state it neutrally.

    I would expect you and your site’s posts to take a similar approach. Strive for neutrality, accurately informing the public, and trying to encourage improvements in the Census by pointing out things that need to be improved. If you come across legitimate ‘gotchas’ in the process, report them accurately. But don’t inflate, fabricate, or carelessly-imagine ‘gotchas’ just to try to make a name for yourself by grabbing attention. Clearly admit when you’re wrong and apologize, and try harder not to make similar mistakes again (and try not to make them in the first place). Continue working to build yourself a solid reputation for quality journalism, instead of periodically devolving into a ‘throw everything at the wall and see what sticks’ approach. Too much thrown at the wall that doesn’t stick will gradually cause your posts about legitimate concerns to be increasingly disregarded.

    I understand your motivation for, perhaps at least sometimes unintentionally, tending to do that (grasp for anything flashy you can kind-of find, and put substantiation to the wayside). It is challenging for young people to get ahead in many careers nowadays, particularly journalism, and you want to make a name for yourself. But you would be better off in the long run, and better serve the stated intent of this website (a ‘watchdog’ isn’t much use if it barks at random leaves blowing by), by making a perhaps-less-flashy good name for yourself for solid journalistic efforts emphasizing reliability, trustworthiness, and accuracy.

    From likely this site’s most loyal reader/commenter, I hope you will understand what I’ve been saying here, and take this advice seriously in the constructive ‘big-picture’ spirit in which it is intended.

  27. Anonymous CL Says:

    Clarification on this paragraph I wrote:

    “Yes, as another Pew article says, “The Census Bureau has not released data showing a comparable figure for the mail participation rate at this point in time during the 2000 Census.” And it would be good if they did release comparable figures. They could put the data into that Take 10 Map application so that you could toggle between a 2000 Participation Rate layer, a 2010 Participation Rate layer, and comparison layer (is 2010 ahead or behind 2000), each day.”

    They do currently have 2000′s final Participation Rates available in the Take 10 Map, but not the day-by-day equivalent Participation Rates from 2000 to compare to the 2010-so-far Participation Rates; and not as an alternate color layer to see multiple areas’ 2000 data simultaneously, only when a specific area’s details are selected.

  28. Anonymous CL Says:

    And another clarification (hit Submit too quick)… you can compare 2000 final Participation Rates and 2010 current-so-far Participation Rates for up to 5 areas at a time, by adding them to the ‘Compare’ graph. Not bad, but room for more capabilities, and definitely room for day-by-day 2000 data.

  29. Dairyland CL Says:

    I have a scary question. I live in a small Wisconsin town and work for the Census, yet I have not gotten anything in the mail from the Census Bureau. I have heard the same thing from several other people. Whether we live in town or on a rural route, we should have gotten something. Also, one of my old crew, like others, has been canvassing homes. What is up with this?

  30. Stephen Robert Morse Says:

    You should report that to the Census Bureau — your bosses — immediately and contact one of the national assistance centers. SRM

  31. Stephen Robert Morse Says:

    Anonymous CL – I really, really, appreciate you taking the time to create such in-depth commentary. Shoot me an e-mail please. SRM

  32. Dairyland CL Says:

    How will this help the under-response situation? I just found from one of my bosses that the entire northern part of Wisconsin has no census questionnaires coming in the mail. Some 12 – 16 counties with the exception of the city of Superior will all be visited by enumerators. That has already begun. Way to save money.

  33. GS-X Says:

    Dairyland CL,

    Sounds as if those 12-16 counties are Update/Enumerate or Update/Leave Type of Enumeration (TEA).
    As many problems as there are with the 2010 Census, I don’t think this is one of them.
    However, you have my permission to call it a failure of the 2010 Census Communications Program.

  34. My Two Census » Blog Archive » MyTwoCensus Editorial: The Census Bureau PR Machine is at it again…Return rates for 2010 are not better than return rates for 2000, and here’s why Says:

    [...] campaign) would exceed the 2010 Census rates. In a comment posted on this blog on March 25, 2010, Steve Jost, the Census Bureau’s Associate Director of Communications wrote “It is tricky… This is an excellent and true point. (The long and short forms for the 2000 Census can be found [...]