Enter your email address to subscribe to Irregular Times and receive notifications of new posts by email.

Join 360 other subscribers

Irregular Times Newsletters

Click here to subscribe to any or all of our six topical e-mail newsletters:
  1. Social Movement Actions,
  2. Credulity and Faith,
  3. Election News,
  4. This Week in Congress,
  5. Tech Dispatch and
  6. our latest Political Stickers and Such

Contact Us

We can be contacted via retorts@irregulartimes.com

Selection Bias: In Video from 7 Cities, Americans Elect Underrepresents Women and Black People

Political campaigns in the modern era commonly use demographic targeting to conserve resources, appealing to certain portions of the population and largely writing off others:

J. Brian O’Day explains:

The point of targeting is to determine which subsets of the voting population are most likely to be responsive to your candidate and focusing your campaign efforts on these groups of voters…. Demographic targeting is splitting the voting population into various groups or subsets of the population. These groups can be based on age, gender, income, level of education, occupation, ethnic background or any other distinct grouping. The point of breaking the population down like this is that similar people are likely to have similar concerns and vote for the same candidate.

It is often easy to determine which demographic groups you are willing to give to your opponents once you have decided which groups are yours. They are the opposite of the groups that you consider to be your best target groups. For example, older male pensioners are going to have the least in common with young working mothers, so if you have targeted one group, you will most likely give the other group to your opponents. It would be very difficult to develop a message that will persuade both groups that you have their interests at heart.

What is the target demographic of Americans Elect in its bid to arrange the election of its own candidates for President and Vice President of the United States in 2012? Americans Elect has provided the first indication in its release of six promotional videos on its home page and one more on its blog. These videos feature 7 minutes and 8 seconds of excerpted on-the-street interviews with 61 individuals in 7 different American cities.

Below are my tabulations of gender and skin color for people appearing in each of the seven videos. I acknowledge that tabulation of skin tone is somewhat subjective; I encourage you to review these videos for yourself and make your own judgment in that regard. I’ve also included information on the demographic distribution by race and gender as tabulated by the Census Bureau (and easily available for your review via the Census’ American Factfinder).

Baltimore, MD
% black in population: 64%, % black in video: 0%. Black population of Baltimore under-represented by 64%.
% white in population: 30%, % white in video: 100%. White population of Baltimore over-represented by 70%.
% male in population: 47%, % male in video: 50%. Male population of Baltimore proportionally represented.
% female in population: 53%, % female in video: 50%. Female population of Baltimore proportionally represented.

Cleveland, OH
% black in population: 53%, % black in video: 16%. Black population of Cleveland under-represented by 37%.
% white in population: 37%, % white in video: 83%. White population of Cleveland over-represented by 46%.
% male in population: 47%, % male in video: 58%. Male population of Cleveland over-represented by 11%.
% female in population: 53%, % female in video: 42%. Female population of Cleveland under-represented by 11%.

Denver, CO
% black in population: 10%, % black in video: 10%. Black population of Denver proportionally represented.
% white in population: 69%, % white in video: 90%. White population of Denver over-represented by 21%.
% male in population: 51%, % male in video: 80%. Male population of Denver over-represented by 29%.
% female in population: 49%, % female in video: 20%. Female population of Denver under-represented by 29%.

Minneapolis, MN
% black in population: 19%, % black in video: 0%. Black population of Minneapolis under-represented by 19%.
% white in population: 64%, % white in video: 100%. White population of Minneapolis over-represented by 36%.
% male in population: 50%, % male in video: 86%. Male population of Minneapolis over-represented by 36%.
% female in population: 50%, % female in video: 14%. Female population of Minneapolis under-represented by 36%.

Raleigh, NC
% black in population: 30%, % black in video: 0%. Black population of Raleigh under-represented by 30%.
% white in population: 58%, % white in video: 100%. White population of Raleigh over-represented by 42%.
% male in population: 49%, % male in video: 89%. Male population of Raleigh over-represented by 40%.
% female in population: 51%, % female in video: 11%. Female population of Raleigh under-represented by 40%.

Seattle, WA
% black in population: 8%, % black in video: 0%. Black population of Seattle under-represented by 8%.
% white in population: 70%, % white in video: 100%. White population of Seattle over-represented by 30%.
% male in population: 50% male. % male in video: 86%. Male population of Seattle over-represented by 36%.
% female in population: 50% female. % female in video: 14%. Female population of Seattle under-represented by 14%.

St. Louis, MO
% black in population: 44%, % black in video: 0%. Black population of Baltimore under-represented by 44%.
% white in population: 49%, % white in video: 100%. White population of Baltimore over-represented by 51%.
% male in population: 47%. % male in video: 87%. Male population of Baltimore over-represented by 40%.
% female in population: 53%. % female in video: 13%. Female population of Baltimore under-represented by 40%.

Black people and women are often under-represented relative to their presence within their communities in these Americans Elect promotional videos, and this under-representation is often by a large margin. White people and men, on the other hand, are never under-represented in any video, and are often over-represented by large margins.

These differences reflect some form of selection bias, which may have occurred in a number of different ways. It is possible that Americans Elect videographers chose to film in places within these cities more often frequented by white people and men. It is possible that when Americans Elect videographers encountered people on the street, they were less likely to approach them and ask them questions if they were black or female. It is possible that when sorting through raw video footage, Americans Elect editors disproportionately chose male and white speakers to include in the final video. In each case, Americans Elect would have made a choice that (intentionally or unintentionally) led to biased results.

It is not safe to conclude that Americans Elect meant to disproportionately select men and white people to appear in their interview videos from these seven videos, but it is exceedingly unlikely that this pattern would have appeared as a matter of random chance alone. If Americans Elect has selected a target demographic, as most modern political campaigns do, the pattern in the videos suggests an attraction toward white people and men, and a repulsion away from black people and women.

1 comment to Selection Bias: In Video from 7 Cities, Americans Elect Underrepresents Women and Black People

  • Tom

    Now you see that correlation thingie going on again: Republicans don’t like anyone other than lillywhite (and rich) and have an undeclared war on women going on right now too, by their actions/legislation/make-up of the party. Not surprising.

Leave a Reply

  

  

  

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>