Tuesday, March 20, 2007

Mastermind and Interactions

The board game, Mastermind, can be used to demonstrate interactions and main effects in an abstract way.

Here is Wikipedia's description of Mastermind (click here). Downloadable versions of Mastermind can be found near the bottom of that page. Finding original games is still possible. Try an eBay search for Mastermind, for instance.

Playing Mastermind is, ultimately, a search for the interaction of its two variables (a.k.a. main effects) peg color and peg position. A player wins when the opponent's pattern of pegs has been decoded. During play, the coding player provides feedback (or, values for a dependent variable) using white or black pegs. A white peg indicates that the correct color of a peg has been decoded, but that the peg is not in the correct position. A black peg indicates that both the color of the peg and its position are correct. The decoding player wins when all of the feedback pegs are black.

Mastermind can also be used to illustrate (again, in an abstract way) effective experimental design. Decoding players who do not plan their decoding stategy will nearly always fail to discover the coded pattern. Try a game or two.

NY Times article: Tracing the cigarette's path from sexy to deadly

Here's a link to a New York Times article (2007-03-20): Tracing the cigarette's path from sexy to deadly. It details how long it took people and society to realize that cigarettes are dangerous.

National Irag Survey

ABC News, in conjuction with BBC and ARD TV (Germany) has conducted another opinion survey in Iraq using a random sample. Here is how they selected respondents (see http://abcnews.go.com/US/story?id=2954886&page=1 for complete details):

"Four hundred and fifty-eight sampling points were distributed proportionate to population size in each of Iraq's 18 provinces, then in each of the 102 districts within the provinces, then by simple random sampling among Iraq's nearly 11,000 villages or neighborhoods, with urban/rural stratification at each stage."

"Maps or grids were used to select random starting points within each sampling point, with household selection by random interval and within-household selection by the 'next-birthday' method. An average of five interviews were conducted per sampling point. Three of the 458 sampling points were inaccessible for security reasons and were substituted with randomly selected replacements."

One measure of the success of the survey was that none of the interviewers were killed or injured during the course of the research. Comments from the pollsters can be viewed at:

http://abcnews.go.com/US/story?id=2954867&page=1

Some of the main results are summarized below:
  • My own life is going well: 2005/71% said yes and in 2007/39% said yes
  • I expect my life to get better: 2005/64% said yes and in 2007/35% said yes
  • My own neighborhood is not safe: 2005/36% said yes and in 2007/74% said yes
The survey was conducted by D3 Systems of Vienna, VA, USA.