The Census Getting Too Personal?

[bbbamI]

I've said it before and I'll say it again...after I had my first pap, nothing seems too personal to me....

 

[ERD50]

In the meantime, a good use for race data in the census IMO, would be to determine how close we are to race-neutrality, and which direction we are going. Are we getting closer to it or further away? To me, refusing to provide racial data is a sort of denial. It's claiming "as a society, we've already arrived, racism has been completely eliminated, so we don't have to talk about it or do anything about it any more". I can take you to a bus stop not two blocks from my house and prove that ain't true.

I'm not following you, at least as it relates to the census count (which is the topic after all).

Yes, unfortunately racism is not dead. But how does a count do anything to improve the situation? Does learning that we have a 0.1% population of X race tell us how they are being treated, or what help they need specific to their race, or if they even need/want help? To echo samclem's sentiments, wouldn't knowing they were poor and/or illiterate tell us more about how to help them?

Come to think of it, why not a box for "At what grade level can you read/write English"? Read/Write was included on older surveys. I'd guess that has more correlation to who needs help than their race.

For a data-point, my daughter's boyfriend's family can check one of those specific minority boxes. He's at the same highly rated school she is, he's in the Engineering school, his Dad is an engineer with his own business. They seem to be doing pretty well. Yet, we are told to check different boxes. What good comes from that?

I am very tempted to enter something like race; other; human or 'American' - because that is how I feel. I don't see the significance to our governance whether or not my family came from various parts of Europe (as far as I know, which isn't very far). However, if this data is really useful, then I want to fill it out in the most useful way, but it's hard to understand what use it is.

I poked around a little on the Census website, didn't find an answer to the question stated that way.

OK, a little off topic, but... my real fear is that we have already hit the best we are going to get in terms of 'race-neutrality', and we are turning the other way. I sure hope not.

-ERD50

 

[ERD50]

There are some health issues that affect specific races ...

Sounds like just the thing for the CDC. I bet they could learn more from hospital records than from a census.

But again, if there are good reasons for it, I support it. But I'm very skeptical.


Hey, wait a minute, there are some health issues that are specific to groups not included in that survey:

Description: Canavan disease (CD), which primarily affects children of eastern and central European Jewish (Ashkenazi) descent, is an inherited neurological disorder in which the brain deteriorates.

But they are told to enter "White", while others get specific down 'Cambodian" versus "Vietnamese". So no, sorry, not buying that explanation as it stands.

-ERD50

 

[haha]

And, why no breakdown of 'white'? A Serb is the same as a Croatian as a German as a Ukrainian?

They all look alike to me.

 

[Bestwifeever]

Here's the answer to your prayers:

For complaints/concerns about the​

2010 Census and other, operations
write or e-mail to:
Associate Director for Administration
Paperwork Project #OMB-0607-0919-C
U. S. Census Bureau
Washington, DC 20233-7100
<http://www.Paperwork@census.gov>​

Type “Paperwork 0607- 0919-C” in the subject line

 

[kyounge1956]

Yes, I think in many respects we have arrived. There are absolutely no more legal barriers or restrictions based on race. The only place current laws mention race is in the very unfortunate racial references in some programs that are only serving to perpetuate racism.

Now, that's not to say that there are no bigots and racists out there. It appears to me that white racists are, to a large degree, older people, so that problem may be on its way out. I don't know if the same situation (more racial tolerance and "color blindness" among the young than among the old) exists in black communities, I think perhaps it does not. If true, that's unfortunate, and will set us back decades.

Our big cultural "divide" now is along the wealth axis, not among racial groups. Poor black people have fewer opportunities almost exclusively because they are poor, not because they are black. If we can concentrate on the right problem and not get distracted by skin tone, we'll be on the right track.

I'm not saying you are wrong, in fact I think you are right or anyway closer to right than would have been the case twenty or thirty or forty years ago. But how can your assertion be backed up with any kind of valid statistical analysis unless race data is gathered along with economic data? How would it be possible, for example, to tell that poor blacks aren't any worse off than poor whites, or that upper-income blacks are more similar to upper income whites than to lower income blacks, or that a higher proportion of blacks graduate from college now than in the past, or earn above median incomes, without any racial data? Or if those statements I just made aren't true, how could anyone tell if there is no race data? And the same goes for other ethnic groups as well.

How else can we tell whether, as a nation, we are moving away from our racist past, marking time, or losing ground? Without data, all we have is anecdotal observations, and what I observe in the Northwest is possibly very different from what someone else would observe in other parts of the country.

 

[kyounge1956]

I'm not following you, at least as it relates to the census count (which is the topic after all).

-ERD50

What specifically didn't you follow? I will endeavor to explain more clearly.

 

[Glo]

bbbamI: LOL, my feelings exactly!!!

 

[samclem]

. . . But how can your assertion be backed up with any kind of valid statistical analysis unless race data is gathered along with economic data?

Good point--how do we know there's been progress without metrics? Two observations:
1. I'd much prefer that the racial questioning not be done by the government. Let the Serbs and Croats tally by ethnicity, every citizen in the US, as far as the government is concerned, is an American. Period.
2. Do these stats on race and economics tell us much useful? Blacks are poorer than whites, in general. Does that give us information that is useful? I'm much more interested in the overall distribution of income and how to give poor people an opportunity to improve their lot. Whether they are black, white, brown, etc--how does that helps us with the underlying issue of poverty?

 

[ERD50]

this was directed to me (darn 'no embedded quoting' option!):

What specifically didn't you follow? I will endeavor to explain more clearly.

Let me use your next post as a springboard to answer:

But how can your assertion be backed up with any kind of valid statistical analysis unless race data is gathered along with economic data? How would it be possible, for example, to tell that poor blacks aren't any worse off than poor whites ...

My form has no questions regarding economic situation (other than a general housing question - own w/mortgage, own free& clear, rent, pay no rent - no $ figures).

So this is why I don't follow you. It does not tie economic data to 'race' (whatever that means anyhow, but that's another story?).

Plus, as we said before, why aren't these detailed subsets of whites listed - perhaps certain subsets of whites are in the same economic category as some of these other subsets, but we would never know since they are all lumped together? I guess that is what gives me the feeling that the form itself is discriminatory. Not in a "I want to jump up and down and turn red and stomp my feet" kind of way, but in a "that sure seems odd, is this really what we should be doing?" kind of way.


-ERD50

 

[kyounge1956]

this was directed to me (darn 'no embedded quoting' option!):

What specifically didn't you follow? I will endeavor to explain more clearly.

I want embedding too! I can fake it using the multi-quote button, but I wish it was just automatic.

Let me use your next post as a springboard to answer:

But how can your assertion be backed up with any kind of valid statistical analysis unless race data is gathered along with economic data? How would it be possible, for example, to tell that poor blacks aren't any worse off than poor whites...

My form has no questions regarding economic situation (other than a general housing question - own w/mortgage, own free& clear, rent, pay no rent - no $ figures).

So this is why I don't follow you. It does not tie economic data to 'race' (whatever that means anyhow, but that's another story?).

I didn't know there weren't economic questions on the form, because I haven't opened my questionnaire yet. (Having read here on the board that the directions are (1) provide answers to the questions as of April 1, and (2) mail the completed form TODAY, ISTM the only way to comply with both is not to open the envelope for another 10 days. And I'm not going to go look at the envelope again to see if it says "Open Immediately" on the outside, because if it does the insoluble dilemma might cause my brain to explode.) So anyway, I didn't know there were no economic questions in the census form, and that's where I was missing your point. If there is a long questionnaire that only some households fill out as there was in 2000, I would imagine the economic questions are in there, but if not, I'm at a loss. The census data is also used for defining Congressional districts and possibly the race data would also have some applicability there. I have a vague recollection that there were some Supreme Court decisions wrt racial discrimination by gerrymandering The race data might be needed to demonstrate compliance with the Court's decision.

Plus, as we said before, why aren't these detailed subsets of whites listed - perhaps certain subsets of whites are in the same economic category as some of these other subsets, but we would never know since they are all lumped together? I guess that is what gives me the feeling that the form itself is discriminatory. Not in a "I want to jump up and down and turn red and stomp my feet" kind of way, but in a "that sure seems odd, is this really what we should be doing?" kind of way.


-ERD50

I agree it doesn't make much sense to have one ethnic group defined very broadly while another one is divided into many subsets. I have no idea how it was determined which groups would get a checkbox on the form, but I would guess that when the data gets back to HQ, there will be considerable lumping-together of subsets into broader categories. I also think it would make sense for "declines to state" to be one of the choices under Race. I think a count of people who don't wish to answer would make the data more accurate on the whole than if people lie or make up answers. It would still probably be possible to estimate the ethnic makeup of the population as a whole, and to lay out Congressional districts fairly, and if the number of Americans who don't identify themselves racially rises over time, that might also be an indicator of movement in the direction of race-neutrality.

 

[kyounge1956]

Good point--how do we know there's been progress without metrics? Two observations:
1. I'd much prefer that the racial questioning not be done by the government. Let the Serbs and Croats tally by ethnicity, every citizen in the US, as far as the government is concerned, is an American. Period.

How would you gather the metrics then? Make the census strictly a head-count and then have some other body (university?) get the details on a statistically valid sample that could be extrapolated for the entire population? That might work...

2. Do these stats on race and economics tell us much useful? Blacks are poorer than whites, in general. Does that give us information that is useful? I'm much more interested in the overall distribution of income and how to give poor people an opportunity to improve their lot. Whether they are black, white, brown, etc--how does that helps us with the underlying issue of poverty?

Even assuming for purposes of discussion that poverty is in fact the mainl problem and race an insignificant factor, I think it's possible that, due to cultural differences between ethnic groups, the same government program might for example be effective in helping members of group A escape poverty, have no effect either way on members of group B, and be a hindrance to people in group C; poverty in group D might come from different historical causes, requiring different remedies, than poverty in group E; a program might be presented in such a way that people in group F avail themselves of it and benefit, while people in group G don't, and so on. It is analogous in a way to the medical tests that are now being developed can tell by the presence or absence of a gene marker whether a medication will be effective for a particular patient. But I think the only way to tell one way or the other is to collect data including race, along with other information.

But as ERD has pointed out, the census may not actually be gathering the data necessary to do this sort of analysis, in which case we'll be doing everything by guess and by golly for the next ten years.

 

[Texas Proud]

How would you gather the metrics then? Make the census strictly a head-count and then have some other body (university?) get the details on a statistically valid sample that could be extrapolated for the entire population? That might work...
Even assuming for purposes of discussion that poverty is in fact the mainl problem and race an insignificant factor, I think it's possible that, due to cultural differences between ethnic groups, the same government program might for example be effective in helping members of group A escape poverty, have no effect either way on members of group B, and be a hindrance to people in group C; poverty in group D might come from different historical causes, requiring different remedies, than poverty in group E; a program might be presented in such a way that people in group F avail themselves of it and benefit, while people in group G don't, and so on. It is analogous in a way to the medical tests that are now being developed can tell by the presence or absence of a gene marker whether a medication will be effective for a particular patient. But I think the only way to tell one way or the other is to collect data including race, along with other information.

But as ERD has pointed out, the census may not actually be gathering the data necessary to do this sort of analysis, in which case we'll be doing everything by guess and by golly for the next ten years.

The census bureau gets a lot of information in other ways.. in fact, I had heard, read, something a long time ago that they had hoped they did not have to do the 'count everybody' every 10 years and that number is 'flawed' and they can not correct it...

FWIR, they can do statistical sampling and get a better number with better data for a lot less money... that was part of the fight they have with Congress... Congress repubs want a 'true' count as it seems minorities do not fill out the form or do not fill it out correctly (do not remember)... or they are living in a place that is not counted... the census wanted to 'fix' the numbers, but Congress said 'no'...

SOOO, they can get all their racial numbers, income numbers, etc. etc. with their sampling... when you get the long form... you are part of the sample...

 

[ERD50]

I didn't know there weren't economic questions on the form, because I haven't opened my questionnaire yet. (Having read here on the board that the directions are (1) provide answers to the questions as of April 1, and (2) mail the completed form TODAY, ISTM the only way to comply with both is not to open the envelope for another 10 days. And I'm not going to go look at the envelope again to see if it says "Open Immediately" on the outside, because if it does the insoluble dilemma might cause my brain to explode.)

You made me look! Yes, inside it does say "Please complete and mail back the enclosed census form today". Darn, now I am stuck in that conundrum, shouldn't have opened it until 4/1. You have been warned!

I have a vague recollection that there were some Supreme Court decisions wrt racial discrimination by gerrymandering The race data might be needed to demonstrate compliance with the Court's decision.

That's a possibility - in my view though, gerrymandering for political advantage is the problem - that can be done based on factors other than race that help keep the incumbent in power rather than serve the public. I'm a 'root problem' kind of guy - let's fix gerrymandering, and then all gerrymandering problems are fixed.

and if the number of Americans who don't identify themselves racially rises over time, that might also be an indicator of movement in the direction of race-neutrality.

Interesting thought - it got me thinking - let's roll with it and maybe take in a slightly different direction (assuming we think there should be any q's like this at all on the census...). What if instead of just "declines to state", what if it was more specific, like "I don't believe that my racial background is of significance to the Federal Govt"? When 100% of the returns come back that way, I think it tells us that the Fed Govt should move on to other problems.

I want embedding too! I can fake it using the multi-quote button, but I wish it was just automatic.

Call your [-]Congressperson[/-], ummm forum administrator! Make your voice heard! IIRC, this forum used to support it, and other forums I'm at do. You do not want it to be automatic with every post quoting everything, but if you do choose to quote, it sure would be nice to have the quoted quotes included. I can also 'fake' it but it gets tedious and the formatting sometimes gets screwed up.


Thanks for your thoughts on the subject (all of them) - ERD50

 

[MichaelB]

The census data is also used for defining Congressional districts and possibly the race data would also have some applicability there. I have a vague recollection that there were some Supreme Court decisions wrt racial discrimination by gerrymandering

The race data might be needed to demonstrate compliance with the Court's decision.

Population count and location are needed to draw congressional districts – for local, state and national governments.

Race is not to categorize the population, it is needed to draw districts that ensure minorities are not denied representation by gerrymandering – a sort of reverse segregation. Race categories are driven by this need, so they represent a listing of minorities that might need some type of current or future districting at a local or state level. A lot of that is based on local and state requirements.

The census bureau gets a lot of information in other ways

They do (it does?) – which is why they dropped the long form. The census bureau is continually doing surveys and projections. This is just one – albeit an unusually big one...

 

[samclem]

That's a possibility - in my view though, gerrymandering for political advantage is the problem - that can be done based on factors other than race that help keep the incumbent in power rather than serve the public. I'm a 'root problem' kind of guy - let's fix gerrymandering, and then all gerrymandering problems are fixed.

Let's put aside real-world political considerations for a moment. Wouldn't it be great if congressional districts were constructed using some completely objective method? Maybe based on MGRS grid-squares or other means that simply attempts to equalize the populations in each district while striving for optimum "compactness." It would take all the politics and gerrymandering out of the equation.

Okay, now back to the real world . . .

 

[kyounge1956]

Let's put aside real-world political considerations for a moment. Wouldn't it be great if congressional districts were constructed using some completely objective method? Maybe based on MGRS grid-squares or other means that simply attempts to equalize the populations in each district while striving for optimum "compactness." It would take all the politics and gerrymandering out of the equation.

Okay, now back to the real world . . .

I don't think so, because back here in the real world, the geographic distribution of ethnic groups isn't random. Districts of optimum compactness might still result in unevenness of political "clout" between one group and another, by (for example) splitting up a concentrated population of one ethnic group into several districts, putting them in the minority in each, instead of being in the majority in one district and barely present in the others. I think in the past instances of racial discrimination by gerrymandering this was done deliberately, but it could still happen accidentally if race is ignored in drawing district boundaries.

I am afraid optimally compact Congressional districts will just have to wait until the Census race data shows random distribution of each ethnic group over the entire country, or at least each state.

 

[samclem]

. . . but it could still happen accidentally if race is ignored in drawing district boundaries.

If it happens accidentally then I wouldn't call it gerrymandering--I'd call it "bad luck."

And, if we're culling the census data to assure various ethnic and "racial" groups are treated fairly (which, I guess, means that the expected election results will produce the "right" number of elected representatives for each group), then I think we're way off base. Are we doing the same thing for gays? For left-handed people? There's a big difference in expected lifetime income for short people vs tall ones--are we assuring short people get the "right" representation? How about something as fundamental as income level? Religion? I can assure you that atheists are not elected to public office in anything like their prevalence in the population. There are >>many<< other factors far more important than race, and the more we obsess about skin color, the more we slip into the "racial spoils" game. Nope--it's time to stop asking or answering any questions regarding race or ethnicity on the census or any other government form.

BTW, if, as expected, Republicans gain a lot of seats in the next election, they'll be in a position to lock that advantage in as they influence the once-a-decade redistricting process. What goes around comes around.

 

[ERD50]

I don't think so, because back here in the real world, the geographic distribution of ethnic groups isn't random.

....

I think in the past instances of racial discrimination by gerrymandering this was done deliberately, but it could still happen accidentally if race is ignored in drawing district boundaries.

What samclem said, plus - I'd fully expect that randomness would even out political clout on average, far more effectively than politicians attempting to 'get it right' by some 'process'. Plus, it would eliminate the 'on purpose' discrimination.

-ERD50

 

[M Paquette]

I want embedding too!

Embedding? Is that the trick where the census taker identifies himself as a cute kid named Cousin Oliver, and then lives as part of your household while secretly filing census reports?

 

[kyounge1956]

If it happens accidentally then I wouldn't call it gerrymandering--I'd call it "bad luck."

And, if we're culling the census data to assure various ethnic and "racial" groups are treated fairly (which, I guess, means that the expected election results will produce the "right" number of elected representatives for each group), then I think we're way off base. Are we doing the same thing for gays? For left-handed people? There's a big difference in expected lifetime income for short people vs tall ones--are we assuring short people get the "right" representation? How about something as fundamental as income level? Religion? I can assure you that atheists are not elected to public office in anything like their prevalence in the population. There are >>many<< other factors far more important than race, and the more we obsess about skin color, the more we slip into the "racial spoils" game. Nope--it's time to stop asking or answering any questions regarding race or ethnicity on the census or any other government form.

BTW, if, as expected, Republicans gain a lot of seats in the next election, they'll be in a position to lock that advantage in as they influence the once-a-decade redistricting process. What goes around comes around.

I don't define "treated fairly" as ensuring that the percentage of any given ethnic group in Congress is equal to that group's percentage of the population. If that were my definition, I would be screaming bloody murder about under-representation of women—I think Congress is still much less than 50% female.

To me it doesn't matter if the denial of proportionate clout is deliberate or accidental—you still have one group gaining political influence to the detriment of another. This doesn't just apply to racial groups. For example, it would be possible to divide my state (and perhaps many others) into equally populous districts, in each of which the urbanites outnumber the rural/small town residents, creating a situation in which perhaps a quarter or a third of the population of the state live in small towns and rural areas, but all the representatives are determined by urban votes. A candidate in such a state could well afford to ignore rural concerns, because s/he can get elected by getting a large majority of votes in the city only, but can win 100% of the rural/small town vote and still lose the election. The same might apply to high & low income levels, if the geographical distribution of wealth is non-random enough to make high- or low-income people vulnerable to gerrymandering as other segments of the population are. My concern doesn't apply to southpaws or short people or atheists, because, to the best of my knowledge, those groups are randomly distributed geographically. No amount of accidental or deliberate gerrymandering can affect them, because they aren't highly concentrated in a single geographical area. Gays are a different case. I would be surprised to find that homosexuals are randomly distributed across the entire country. I would not be surprised if there are at least a few legislative districts (e.g. in San Francisco?) in which gays are sufficiently numerous that their votes can swing an election one way or the other, and they can't be ignored with impunity as could the ruralites in my other example.

You wrote, 'If it happens accidentally then I wouldn't call it gerrymandering--I'd call it "bad luck.' I might be willing to go along with that if there is a mechanism in place to prevent these accidental unfairnesses from becoming permanent. If, in addition to being "optimally compact" all district boundaries have to be re-drawn in such a way that the new districts contain parts of at least two or three old districts, or such that some predetermined percentage of the population must end up in a different district than they were in previously, it might ensure that any such "bad luck" lasts at most ten years. I would like to see the same principle applied to party affiliation too, so that neither party is a shoo-in in any district, nor can either party's candidate afford to ignore the concerns of members of the other party. Maybe that would promote a less-polarized political atmosphere.

 

[haha]

Whether they are black, white, brown, etc--how does that helps us with the underlying issue of poverty?

I suspect that you know the answer your question. It doesn't help with poverty, and furthermore it is not intended to.

Do you know how many political and bureaucratic and business careers have been created by the elaborations of minority politics? For memebers of all races?

None of these people want to lose power, influence, or employment, and since this is the US, they will not.

Any visible improvement would tend to erase the "condition" that requires all this activity, so any improvements have to be hidden or denied.

Minority politics is like the national debt, it gets rolled over, it grows, the one thing it doesn't do is become smaller.

And our job is to shut up and pay the bills.

And to answer the narrow question of why these race questions are found on the census- how about congressional redistricting?

Ha

 

[MichaelB]

Wouldn't it be great if congressional districts were constructed using some completely objective method?

You got that right. Along with election rule reform, significant change in these two areas would enable major changes in our we govern ourselves. Not likely, but I would really enjoy discussing it with you over beers someday.We can invite some of the other active opinionators on this thread and have an enjoyable moment.

District design is done by state legislatures, the big changes would come from states that will [-]suffer[/-] [-]enjoy[/-] register changes in legislative majority between 2000 and 2010. There may not be many.

Gerrymandering is a problem because it has allowed both political parties to draw districts that concentrate their own supporters, reducing the likelihood of being voted out of office by opposing party candidates. This is why turnover is so low in state and federal districts and also why the real battles are in the primaries. It also explains why both parties tend to push toward their extremes.

Race is used two ways - to ensure minorities are not denied representation and also to enable representation. As Ha points out, once established, a district it is not likely to disappear. Still, it is up to the initiative of the individual state.

I found the predominance of asian categories unusual and wonder if it reflects an effort to increase focus on asian political representation.

 

[Elderdude]

I overheard a blurb on some talk radio show about the census answers being used during WWI and WWII to round up Japanese and/or Germans. I googled this and found it is true for WWII. Census information was also used as recently as after 9/11 for keeping an eye on Arab Americans. I'm going to stop searching now. I'm getting paranoid.

 

[vlagirl]

It has always been my understanding that the only information you are legally required to give on the census is the number of people in your household. I answered question #1 and that's it. If they want to fine me, bring it on.