Linn County, Missouri, population 13,754, is in the north-central part of the state. According to this encouraging article in the Linn County News, a large majority of respondents to its recent reader poll oppose racial and gender preferences.
Seventy-one percent of LCL readers responding to the poll “strongly disagree” with the statement “as a hiring and college admissions policy that gives preference to women and applicants of color, the use of affirmative action should be continued.”
Twelve percent “disagree” with that statement, 12 percent were “not sure,” and six percent “strongly agree.”
So, the percentage who either “disagreed” or “strongly disagreed,” i.e., who oppose racial and gender preferences, was 83%.
The editor pointed out that “the Linn County Leader received 21 responses to this question,” and thus it would be a mistake to read too much into these results. But it would also be a mistake to read too little into them.
the fact that it’s just 21 people is bad, but the real problem is that it was a self-selected group of respondents rather than a random sample of the county. the results tell you about the 21 people who responded and don’t allow any inference to the broader population of Linn.
21 responses from a biased sample? Sorry, John, but the less we read into these results the better.
Also, the numbers don’t make sense. 6 percent “strongly agree”? One person would be 5 percent and two people would be 10 percent, so how many is 6 percent?
John, is it just me, or did the Linn County Leader make some basic math mistake? 12 percent of 21 is 2.5; how can 2.5 people choose “disagree”? If two people chose “disagree”, the percentage is 9.5% (which could be reported as 9% or 10%, I guess); if three people answered that way, the percentage is 14% (rounded from 14.3%).
Based on the reported percentages, 15 of 21 people chose “strongly disagree” (=71.4%). How the other 6 people polled is a bit of a mystery.
Well, in all fairness 21 is a better sampling percentage of 13,754 than is one or a few thousand of 330,000,000. I’ve never seen a national poll that used a sample greater than a few thousand.
But yeah, I wouldn’t order the celebratory Boones Farms Strawberry Hill yet.
the sample size problem is pretty big and obvious. but you guys are missing another pretty obvious point: how the question was phrased. i suspect that if the LCL readers were asked to asses this statement: “as a hiring and college admissions policy which promotes diversity in education, the use of affirmative action should be continued,” certainly less than 83% would oppose it. if the MCRI stuff taught us anything, it’s that a simple twist of language can go a long way in getting people to support certain ideas.
True, Chauncey, but then again so is using the term affirmative action in place of racial discrimination. AA can mean many things, many of which the CRI would support. But put it in plain language, as in “the use of racial discrimination to increase diversity” (and that is plain language, including the “good intentions” of diversity, or at least skin color diversity) and I bet the numbers would shift again.
The point is that telling people that racial discrimination is not only acceptable but a moral imperative doesn’t pass the smell test with generations raised to see it as evil.
mikem —
It’s the size of the sample that matters, not the fraction of the population being sampled.
For example, say that you randomly sample 1 person out of 10 for party affiliation. If that person is a democrat, would you have any good reason to believe that most of the others are democrats? Now consider randomly sampling 100 persons out of 1 billion. If they are all democrats, you can be pretty sure that most of the rest are as well.
Thanks, LTEC. I know that 26 is too low a number to draw firm conclusions from. I was just, sort of, being cute.
I personally think, in fact I know from observing the actual value given to diversity by “diversity advocates”, that it is just an excuse to elevate preferred races over others. It is selectively valued, applied and demanded depending on which race or (sometimes) political ideology will get an artificial boost from it.