With help from his super smart friends, Robert Santos, President-Elect of the American Statistical Association, internationally recognized expert in the census, and Chief Methodologist at the Urban Institute, answers all your questions about why the 2020 census is so important.
So, what’s the census anyway? It’s when the federal government counts every resident in the US every ten years (Yep, every last one). Who should care about the census? Um, everyone. Why should everyone care so much? For sooooo many reasons. It’s required by Constitution, for starters! It sends BILLIONS of $$$ around the country, too (Yup, you read that right. Billions of dollars…). And if that’s not enough, the census determines how many seats your state gets in the House of Representatives. OMG!
Ask Robert and his friends listed below all your questions about the 2020 Census so we can talk about why it’s such a big deal (answers initialed by experts). And then continue the convo by requesting to join the CountOnStats LinkedIn group and by following @CountOnStats on twitter…
2020 Census resources: https://2020census.gov/en
3:30pm UPDATE: Hey Reddit, thanks for all your questions today! We've got just a few more minutes left today to geek out on the census, but we'll be watching this thread closely and will return to answer pressing questions!
11/15/19 3:55pm eastern time UPDATE: Hey Reddit, we're signing off for now, but we'll be monitoring this thread to answer pressing questions. Thanks for the questions!
Hi Rob! Is missing data an issue for census? How do you deal with that issue?
you bet. All decennial censuses experience missing data. There are two types!
1) some questions on the form are not answered
2) No completed form is returned (ie everything is missing)
Guess what? When some Qs are not answered (ie, 1 above), sometimes an enumerator will contact you/the household and ask for that info. And sooooometimes, the Census bureau will fill it in with a statistical prediction!
And when no one returns a form and an enumerator is unsuccessful in getting a response from you *or* your neighbor (yah, called a proxy response), if the Census Bureau knows someone lives in the uncooperative house, they will *impute* (predict/fabricate statistically) the entire form! That happened 6M times in 2010 and it will be much more IMHO in 2020
In your opinion, what could be done to improve statistical literacy in America?
-me, a former math teacher who gave the hell up and moved to El Salvador
I think the only way to increase statistical literacy is by moving away from the formulas and calculations (which definitely have their place, but when it comes to the general public it's further down the line) and focusing first and foremost on the role of statistics in problem solving and story telling. Statistics can really help to provide texture to an issue like, say, food insecurity (people/families that go hungry). Food insecurity is pervasive and survey stats show it. But you can't really solve food insecurity (as "in the long term") by pushing out emergency food. You need to understand the reasons behind it -- reasons that statistics can provide: food insecurity happens in households with chronically ill or disabled people, the elderly with little or no income, families who have a terrible time paying the rent or utilities or medicine, folks who cannot find a job... Statistics serve to tell different stories of the types of people suffering from food insecurity and the solutions need to attend to peoples' specific situations. Bottom line: infuse stats as a natural part of the critical thinking process, and that would get you more than half way to statistical literacy. IMHO
Given the data and computational landscape has changed drastically over the decades, what are the data privacy concerns when reporting on the census?
Thanks for raising this important topic. The Census Bureau takes this extremely seriously and Title 13 requires strict confidentiality protection for respondents. Basically, the Census Bureau is prohibited from sharing any personally identifiable information with any other governmental agency, business, or any other organization for any reason. There are valid concerns about the possibility of people 'hacking' into the census, but the Census Bureau is working hard to make sure that this doesn't happen. -MM
With electronic storage, filing, and analyzing of census information, should we do a census more frequently than every 10 years? It seems like an outdated timetable now that those tasks can be completed faster.
Great question! The Census Bureau started conducting the American Community Survey (ACS) in 2005 to address this very concern. The ACS--unlike the decennial census--is designed to collect timely demographic, housing, social, and economic data every year. However, we still need the decennial census to count all people living in the United States. -MM
What would you consider to be the biggest gap or failure in the current design of the census?
From Tom Louis: In my view, the biggest gap is, due to funding limitations and other problems, the bureau wasn't able to do as many full-scale field tests as would be desirable. So, there are some uncertainties, but it is likely (not certain!) that no major problems will emerge.
Are there any notable differences in the way the United States carries out its census compared to other countries? Are some countries using better methods? Do some countries have great difficulty in getting an accurate count due to limited resources?
Is there one country that stands out above all else in terms of accuracy and efficieny?
There are notable differences among countries. For example, some countries have continuously updated, residence registries and can get a count almost a the push of a button. However, even in these countries there can be data lags, so it's not perfect.
In your opinion, what data from the US Census needs more visualization? What data is overlooked, and what kind of stories are there to tell from that overlooked data? I’m starting a club at my university that focuses on building clear and compelling data visualizations that effectively communicate data, and one area I want to focus on is the US Census. Do you have any thoughts on what stories you in particular would like to see produced using data from the Census? Thanks!
Good question! The strength of the census data is in the availability of data for very small geographic areas, so it's a great resource for MAPS. The availability of trend data (as far back as 1790) also makes it a great resource for trend analysis, although you need to consider the comparability of variables over time (like racial/ethnic categories, which tend to change in every census).
For example, one interesting story you can tell with census data is the population loss in many counties in the Midwest--some of which have been losing population for decades. -MM
Dear Robert - what’s the best way to learn statistics?
<3<3<3 Thanks for asking!
IMHO the best way learn statistics is to view it through a few lenses:
Recognizing and valuing the many facets of statistics is a first step in learning stats. The key is not to see stats as this "hard thing with formulas" but a tool that allows the uncovering of insights. Keeping that in mind helps make the journey of learning statistics a sweet yellow brick road....
What is something people never ask about but you'd like to tell us?
One of the biggest misunderstandings about Census accuracy is the fact that net undercounts are not the same as people missed.. Net undercounts are a balance of people missed and people double counted... Omissions measure people missed in the census.
How do you deal with demographics that are either less likely to complete their Census or might not trust the Census (whether because of general anti-government feelings or because they're afraid/think the citizenship question is still on it)?
There are many Hard-to-Count (HTC) populations in country who are less likely to self-respond to the Census. The Census Bureau is designing an extensive communication campaign to try to encourage such groups to respond to the Census. Part of that effort is a partnership program that works with groups outside the federal government who are "trusted voices" among the HTC groups, Organizations like the Leadership Conference on Civil and Human rights, the National Association of Latino Elected Officials, the National Urban Leagues and the Partnership for Americas Children and many many other organizations have been working in their communities to try and overcome resistance to filling out the census ..WPO
how has changes in asking about race affected analysis? Are their any examples where collecting race data has provided unmistakable social benefit?
Starting in the 2000 Census people were allowed to mark more than one racial category... That made race more complicated to analyze but also more reflective of reality. If racial data were not collected in the Census (and other federal government data collection activities) it would difficult to prove (or disprove) racial inequities in things like education, employment, and voting rights. One of the issues in the 2020 Census is how to provide data for smaller racial groups ( for example people who are black, white and Asian) while at the same time protecting respondents confidentiality...WPO
What’s your opinion on adding middle eastern as an ethnicity on the census? I’ve heard a few people talk about this and I’m curious to see what y’all think
As you may know, the Census Bureau proposed adding a category of Mid-Eastern/North African (MENA) to the Census race categories to the 2020 Census, but that plan fell through when OMB failed to endorse such a change...that was in late 2017 early 2018. I think it would be a good idea to add such a category because it matches the reality as many people see it... I think a lot of people see folks from this part of the world as a minority group within U.S. society ..WPO
How much would it screw up the census if we all marked that we were Samoan?
Edit: iamasamoan.com is available, just sayin'
hahaha... it wouldn't pass the laugh test! and those responses would not be used (imho)
Seriously, back in the 1970 census a "Hispanic" Q was added that asked (among other categories) if people were "Central American", "South American" etc. But the Q WAS NOT TESTED -- it was just inserted into the Census form. Lo & behold, when responses were being tabulated a bunch of Hispanics were found to live in Alabama and Iowa and those areas. When the census bureau did some digging, turns out that folks in Iowa considered themselves living in the central pare of the USA ("America" so they marked that. And in the southern states, folks figured they lived in the "South" so they were "South American" :-)
At which point in your education did you start specializing on this and why did you do so?
I wanted to be a math prof but my undergrad adviser steered me towards statistics, so I went to UM Ann arbor to a Math Stat grad program! Then almost by accident I was called to a meeting by a researcher at ISR Mich who was doing a first ever national survey of Mexican Americans (which is what I am), and he convinced me in a single hour to become a sampling statistician in survey research. That is how I became an applied survey statistician (back in 1977!). I tell the story in my AAPOR Heritage Interview (I was AAPOR Pres. in 2014): https://www.youtube.com/watch?v=LJf4nGpU2ns
We live in a different world than the one that existed almost three quarters of a century ago. I would eschew making a decision today based on events of the distant past (including Japanese internment, slavery, genocide of Native Americans, prosecution of Mexican immigrants and native born, alike). Instead I would point to the Census Act that stipulates protections of privacy and the high level of scientific, constitutional and ethical integrity of the Census Bureau staff -- a bunch of whom I have know personally for 30+ years. I totally believe that they will protect all US residents.
Having said this, I totally understand and I am sensitive to the fear and distrust spurred by recent events (eg, the citizenship question fracas). We have a right to be somber and cautious. So thank you for expressing that. All I can say is that I really believe that the risk of census data disclosure for federal enforcement purposes is extremely low. To my knowledge, no one worries that the IRS staff or the Social Security staff will pass over private data. Well, the public disclosure protections set by the Census Act are far more stringent! We have so much more to gain and the risks are so teeny that to me it is a no brainer: we ALL deserve to be counted!
Is it a crime to not answer the census?
Yes, but to my knowledge no one has ever been convicted of not participating in a decennial census. It's not about the legal mandate; it's all about the civic obligation to yourself, your family, your community, and your state.
When was the last time you said "wow" in your job?
I do that a lot in my job :-)
The biggest wow I had was when I did a project to get a guesstimate on the possible undercount of the 2020 Census. We superimposed the supposedly great performance of the 2010 Census (which overall estimated virtually a dead on accuracy) on a population projection of the 2020 US population. And we came up with a noticeable undercount! The reason was that our wonderful US population is becoming more diverse over time, and the folks who are harder to count are increasing in size. That was a pretty big wow moment for me. (see https://apps.urban.org/features/2020-census/ for more info)
What your real best guess on the number of undocumented people living in the US?
Jeff Passell at the Pew Research Center has been producing estimates of this population for years. Homeland Security also produces estimates of the undocumented population. Estimates from both of these sources are around 12 million undocumented immigrants in the U.S. Given agreement from both of these sources I think that number is reasonable. Please keep in mind that there are a lot of non-citizens living in the U.S. who are not undocumented people. WPO
It bothers me that census is linked to address, will there ever be a point where census is not linked to address but linked to a person?
For the current uses of Census data it is critical to tie people to a geographic location because census data is used to draw legislative districts and census data are used to distribute billions of dollars in federal assistance to geographic locations. If people were not tied to a physical location the census data could not be used for these functions....WPO
What is your favorite Texas BBQ joint?
Statistically speaking, I have conducted a census of BBQ joints in the Central TX area and my all-time fav is Coopers in Llano TX! Those burnt end brisket tidbits are to die for
What happens when a state loses a congressional seat due to population changes? Who determines which districts get changed or merged or whatever?
Based on census numbers the federal govenrment determines how many seats in congress a state gets. Once it is determined how many seats a state get a it is up to the state to draw congressional districts for those seats. The districts must meet certain standards like the equal population standard (one person- one vote) and the Voting Rights Act laws, but states have a lot of latitude on how to draw those districts. Given the heavy partisan political gerrymandering that has occurred in the past few decades more states have started using commissions or more politically independent bodies to create the districts. ..WPO
(big data question) How is census information collected, stored, analyzed, and accessed?
just a quick response to the first part of your questions in terms of how information is collected. Households can self-respond by internet, by phone, by paper questionnaire. The self response process will start about mid March and last though early May. After that, households that have not self-responded will be visited by a Census enumerator to collect their data. That process will end around the last of July. .WPO
If we were to change the census to a more frequent or near-live count rather than every 10 years, do you think that would affect the representatives in various states and therefore larger elections?
Yes this could certainly affect apportionment but the U.S. Constitution mandates that a census be taken only every 10 years. And given the cost of conducting the census--projected at more than $15 billion in 2020--it would be difficult to implement. -MM
1) How did the Constitution's rule that the government must "count the whole number of persons in each State" (Article 1) -- a power brought up solely in the context of apportioning House seats / electoral votes -- transform into a constitutional authority to ask a wide variety of questions and demand that people answer?
2) Why is it not important to ask how many citizens, as opposed to people present, the United States has? This seems like a very basic question we would like to know, as long as we're asking other things than "how many people".
ANSWER TO Q1: The constitution requires a decennial counting of the nation's residents. The Census Act made it mandatory and put in place safety measures that protect the privacy of individuals when they respond. Virtually every question in the decennial census and the "ongoing long form census" (which is actually a survey called the Amer. Community Survey aka the ACS) has a legislative mandate -- that is, Congress requires information for governance that can be mapped to questions in the decennial census and ACS.
ANSWER TO Q2: We actually get info on citizens through the ACS (noted above).
I'm curious why Arabic people who become citizens here are defined as white?
Good question--There was a proposal to change the racial/ethnic categories so that people could identify as Middle Eastern or North African (as a distinct group separate from "White") but this change was not made for the 2020 Census. Hansi Lo Wang wrote a nice article on this topic for NPR: https://www.npr.org/2018/01/29/581541111/no-middle-eastern-or-north-african-category-on-2020-census-bureau-says -MM
With census results being used to influence Congressional funding and redistricting (which are increasingly partisan and gerrymandered), why shouldn’t offended parties voluntarily sit out the census as a form of protest or civil disobedience?
The biggest reason I give for participating is that we *deserve* to be counted. Why? Because we, our community, our town/city/county really really get the resources and representation we deserve and need to make our neighborhoods, our schools, our cities etc a better place to live.
I hate to mention it but by law every person must participate and be counted. But the truth is that participating is much more important than "obeying the law". We deserve our fair share of resources and political representation. Let's not deny ourselves.
Mr. Santos, your title of Chief Methodologist at Urban Institute has caught my attention, and I have one question that I think you will be uniquely qualified to answer: who is your favorite Teenage Mutant Ninja Turtle? Thank you for your time.
hahaha... that's easy: Rafael
Why? That's the only one I remember! :-P
Before this administration, has the census ever been politicized in the past?
There have been political overtones in past censuses but in my experiences partisan politics have never been as big a part of the Census as seen in the 2020 Cycle. WPO
Hello everyone! On your career path was there anything you regretted not learning earlier or not doing sooner?
One of the things that was lacking from my graduate school education (and I think it is lacking from most garduate programs today) is learing how to communicate to those outside my field.. the media in particular. If we can't tell people what we do and why it is important, it will be difficult to get public support for our work. ..WPO
I've heard the census is no longer going to include <1 year old population data, is this true? If so, why?
Not sure where you heard this.. The census is planning to include persons of any age in the 2020 Census... you may have heard that youngest children were missed at the highest rate of any age group in the 2010 Census. According to Census Bureau research about 2 million young childrne ( udner age 5) were missed in the 2010 Census = WPO
Why do we need the census when every citizen already has a Social Security Number? Couldn’t we just count those by the address associated with each?
SSN gets you more than halfway there, but not all the way. Some people "borrow" others' SSNs; others plain out steal them and use them for themselves or sell them. And believe it or not, administrative records are not known for their uniform accuracy -- mistakes can and do happen. Then there is the reconciliation of births & deaths with SSNs. States manage their own the birth/death/marriage tracking systems. (And lots of people move from one state to another, to make matters worse. Have you ever tried to reconcile a state data base with a federal one? It ain't pretty. Bottom line: seems great in theory but when the rubber hits the road, it's always more complex & troublesome.
Why is the general public so bad at statistics? Do you think it’s just not taught well or it’s confusing or something else? It seems like your average person is so wildly off-base when it comes to even basic statistics and probability
I dunno. I think that our society has somehow built an educational/learning system that forgot the value of experiential learning and learning through storytelling. And if we could pull those fundamental learning methods together with critical thinking, combine with illustrations of how statistics can provide insight into everyday problems we have, then I think you'd get "magic".
Why was everyone so up in arms about the census asking if you are a citizen or not?
Isn't the point of a census to get an accurate count of the population for taxes and welfare and stuff like that?
Isn't the point to find out how many US citizens live where?
Another question, what are some interesting population trends you've noticed from census data?
Thanks for the AMA!
Here is a list of some key demographic trends we've identified in a recent report on this topic:
· The U.S. population is increasing and could reach a third of a billion people by 2020, but the population is on track to grow at the slowest rate since the 1930s.
· Fewer households are being established, due in part to the growing share of young adults who still live with their parents.
· As the U.S. population continues to shift to the South and West, states in those regions will likely gain congressional seats at the expense of states in the Northeast and Midwest, changing the political landscape.
· More than half of U.S. counties have experienced net population loss since 2010, with more than 550 counties losing at least 5 percent of their residents.
· The post-2010 demographic situation is especially bleak in Puerto Rico, which has lost more than half a million residents, or 14 percent of its 2010 population.
· The percentage of U.S. residents ages 65 and older is increasing at the fastest pace in U.S. history.
· Children are at the forefront of racial/ethnic change in the United States, creating a diversity gap among generations.
· Growing racial/ethnic diversity in the United States is no longer being driven by immigration, but by patterns of births and deaths among the U.S. resident population.
· A growing divide in homeownership rates between whites and blacks is increasing the wealth gap between racial and ethnic groups.
I have an economist friend working on a high-profile project that uses business-level Census data. How is that collected and what unique challenges does it have compared to the individual side?
Don't know a lot about the economic census but I can say this: It is conducted every five years by the Census Bureau and I would not want to be in charge of that (HIGHLY VALUABLE) monster. Why? Because having designed and conducted establishment surveys, you face increasingly complex issues like peeling an onion. Establishments are born and "die out" at a very rapid rate over time. The *overwhelming* majority of establishments are small businesses. Some don't have addresses. Some change addresses at a rate that is scary. Some are just shell operations so who do you contact? It would be enough to drive me kinda crazy... But hats of the Census Bureau for pulling it off regularly with aplomb...
What business is it of yours how many toilets I have? Why is it my problem if you don't have that information?
My bobcat says hi.
The question about flush toilets was never very popular...and in fact it will not appear on the 2020 census form. It was included on a separate survey questionnaire (American Community Survey) as recently as 2015 but was dropped from that form in order to reduce the burden on respondents. -MM
Everything in your post makes my heart warm. What do you think will be the biggest impact on our fair state (Texas) from the next census?
As a native Texan who loves his state (not necessarily its policies), Texas unfortunately has much to lose. States like California are doling out over $120M on efforts to get decennial participation as high as possible. Sadly, Texas is providing $0... zilch. That means that the risk of an undercount is high and it will predominantly be from areas with high concentrations of Latinos and African Americans. So for the state as a whole we could not receive the number of Congressional House seats that we deserve, nor would we get the federal funding we deserve. Basically, our house seat(s) and funding would got out to other states. And then the story worsens. Within the state, areas in most need and most deserving of federal funding -- such as high concentration Latino ares along the Rio Grande Valley -- would get less than they deserve (because they are most severely undercounted), while more affluent areas with lower concentrations of minority populations would get more than they deserve because allocations within state area a zero-sum game. That goes for planning too: locations of schools, fire stations, roads, legislative districts, or commercial establishments like grocery stores... all are at risk of being placed incorrectly because of an undercount. The stakes are high, which to me means we need a highly charged grass roots effort to convince our beloved brothers & sisters in TX that we *deserve* to be counted!
Your whole ass title was too hard to read, I keep getting confused at statistical association. I have no idea what you do, if I was 4 years old, how would you explain to me what your job is in a much simpler way?
hahaha... even I think my title is too hard to read. I basically try to help people by using numbers creatively to show folks what is good and what is not so good about most anything, from taxes to education to jobs to incarceration to housing, health and so on. Just as a hammer can be used to build a house, a a bridge, a cabinet or a table, statisticians can help build a better car, a better reentry program for ex-offenders, a better delivery system for food banks, better schooling for school kids, etc.
Anything to say about the analysis of privatized data which Caliper published today?
Not sure what you mean by privatized data? If you mean data subject to differential privacy ( the new way the Census Bureau is thinking about protecting confidentiality in the 2020 Census ) there are still a lot of questions about how (and perhaps if) the Census Bureau will use this approach. If they continue down the path they are on, less data will be published in 2020 compared to 2010 and the accuracy of the data may be lower.WPO
The main hiring for the Census Bureau will be in 2020 so I don't think the fact that you were not hired this year is a sign you will not be hired next year. The hiring for 2020 will begin in January with supervisor-type positions but most (roughly 300,000 enumerators )will be hired around April.- May,,.WPO
In your professional opinion, would a citizenship question significantly affect the census results?
It already has. Despite the SCOTUS ruling to leave it off the 2020 census, one cannot ignore the impact of a one+ year media exposure to the public on the citizenship fracas, including the underlying motivation for it in an anti-immigrant policy environment. The damage has been done.
Now, the Census Bureau points to a recent test that showed 'no effect' of adding a citizenship Q. See: https://www.census.gov/newsroom/blogs/random-samplings/2019/10/2019_census_testpre.html
But I point out that this flies in the face of a sizable body of preceding Census and academic research and that the census test cannot be interpreted to mean that in a full decennial census there would be no effect. The test is a matter of "too little too late". See: https://www.urban.org/urban-wire/nonresponse-bias-could-explain-census-bureaus-finding-citizenship-question
hahaha, I love being asked that bc I've mostly been a statistical adviser my career (between jaunts in management & exec service). My best learning came when I realized I should never "just answer the question" when a researcher want "5 minutes" for a quick Q. No I will not spit out a sample size or a power calculation. I want to know what the researcher is trying to learn and why.... what's the basic research question being posed in their study and why is it important. With that as context THEN I digest their question/ask. I found that a noticeable chink of the time, the research question being posed is best addressed with *qualitative research*!! ( which calls for focus groups or in-depth interviews instead of a survey). I love tying together the research Q with the design, the methodology, the data, the analytic plan and the desired insight!
When will the statistical community lead the charge to move away from p-values and towards probabilities?
well... I though we started years ago. check out: " The ASA Statement on p-Values: Context, Process, and Purpose"
Why do you think the citizenship question was not allowed back on the census form and do you think illegal aliens are gaming our system for personal profit?
Well, the facts are that the SCOTUS did not allow the Q and the WH didn't have time to do anything about that (the forms had to be printed). That is why it wasn't allowed back onto the census form.
IMHO -- both documented and undocumented immigrants are too busy trying to legitimately make a life for themselves and their families and stay under the radar in the current anti-immigrant policy environment. Trying to game the system is a great way of getting into legal trouble.
How likely do you think this AMA will be to raise awareness?
Admittedly very little, but by design! Of course there are much more effective ways to raise awareness in the public mind as to the importance of the upcoming 2020 Census (e.g., media campaigns connecting with the advocacy community, CBOs, Complete Count Committees, etc. This AMA is one way to plant seeds of awareness in an already-interested, informed stakeholder community. Hopefully *you* will reach out to others to express your own sentiments on the importance of the 2020 census and the need/opportunity for everyone to participate. That can best occur when folks like you and others who ask Qs participate in these kinds of AMAs.
:-) I was born & raised in TX as were my parents (all four grandparents were born in Mexico). Interestingly, I've never met non-native Texans who understands native connections to our state. Maybe it was the way we all grew up here (e.g, the TX flag, the Texas 'pledge of allegiance', the mix of cultures)
Do I have issue with it? You bet. Texas has a rocky and bloody past. But it's totally home to me. We're a "whole 'nother country"
Why are people against asking if people are legal citizens on the census?
I don't think in the abstract that people are against asking the citizenship question. But asking the Q in the midst of an anti-immigrant policy campaign and the emergence of evidence that the intended use was to undermine our democracy, then people get scared, which in turn raises the risk of **lower census participation***. That is why the civil justice/civil rights communities and many others including myself voice our concerns.
President-elect of the American Statistical Association, What are the odds?
all I can say is that I *am* odd! Guess I'm an outlier...
You think you won the election to be President of the Statistical society, but what are your error bounds?
not sure about error bounds but I *do* know I am bound to error
I think it's unlikely that could happen. The Census Bureau continuously updates a master address file. It's used to send snail mail, deploy field enumerators, and as one component of fraud detection.
TL: The question won't be on the census form so there won't be any `direct' frightening. However, the (failed) effort to include it could well have a general chilling effect.
How can you be proud of living in a place that is so racist, sexist, homicidal and hot?
There exists no place on this earth that is perfect, including TX. And in places where people consider "home" (to me it's TX), we try every day to make our homes better places to live -- through how we treat our neighbors, through civic engagement, through reaching out to folks who need a helping hand. That is what I -- a proud TX statistician --am doing right here in TX. Try it in your home state and see what happens.
Can you say Bonferroni 10x without any errors?
not without a veeerrrry wide berth... ;-)
How can we be sure the Russians won’t interfere in the 2020 census?
Truth be told, we cannot. But we can be pretty sure with the security measures that are in place.
Should I be concerned on how big of an impact differential privacy is going to have on altering census units with low population?
in short: yes :-)
Is it true that, in a liquor store, Methodists will wave to you but Baptists duck?
Statistically speaking, maybe.
When taking the census do you factor in the homeless people? Or are they invisible
The Census Bureau has special procedures to count the homeless, although a big part of that is visiting homeless shelters and service areas during census day (Apr 1, 2020). Here is a link with more info on plans to count the homeless: https://2020census.gov/en/what-is-2020-census/focus/people-experiencing-homelessness.html
Having said all this, some will be missed, of course
What did you think about Canada abolishing the long-form survey a few years ago?
It was a big mistake to make the Canadian long form census *voluntary* and the Canadian gov recognized that almost immediately. In fact, it took 4 years for them to overturn their decision and reinstate mandatory participation (in 2015): https://www.sciencemag.org/news/2015/11/updated-canada-reinstates-mandatory-census-delight-social-scientists
Would you rather fight 1000 hedgehogs or one hedgehog the size of a pig?
Let's not fight, now... give peace a chance. --RS
Hey, how can I get the data from this census, is it public use ?
Question why would anyone give a fuck to ask this joker a question?
hahaha... I ask myself that Q often