Google+
Close

Bench Memos

NRO’s home for judicial news and analysis.

Polling Shenanigans?



Text  



Sunday’s Washington Post will feature this article reporting poll results supposedly showing a “noticeable jump”—from 19% to 31%—in the number of Americans who regard the Supreme Court as “too conservative.”  With respect to two of last term’s cases, the article also reports polling results that strike me as curious.  I expect to have more to say about the poll once the Post makes its data available.  (As I write this, the poll that the article links to doesn’t disclose any results concerning the Supreme Court; I gather those questions are, or are among, the questions 41 through 46 that are listed as “Held for future release.”)

 

Let me highlight now one strange fact that would seem to cast serious doubt on whether the poll was designed and conducted neutrally:  The poll states that it “was conducted … among a random national sample of 1,125 adults, including additional interviews with randomly selected African Americans, for a total of 210 black respondents.”  If I’m correctly understanding this rather confused jargon, black respondents accounted for nearly 19% of the overall respondents.  A U.S. Census Bureau report from 2005 shows that blacks account for about 12% of the U.S. population.  Why, one wonders, did the pollsters include such a massive (over 50%) overrepresentation of black respondents?  (I say that the poll’s jargon is rather confused because it is highly unlikely that a genuinely “random national sample of 1,125 adults” would include 210 black respondents.)


Tags: Whelan


Text  


Subscribe to National Review