Box Turtle Bulletin

Box Turtle BulletinNews, analysis and fact-checking of anti-gay rhetoric
“Now you must raise your children up in a world where that union of man and box turtle is on the same legal footing as man and wife…”
This article can be found at:
Latest Posts

This Month In History: How To Conduct Junk Science

Jim Burroway

February 11th, 2008

Recent election coverage has brought a lot of attention to opinion polls. We’ve seen a number of polls failing to predict some of the state primary winners, while other polls either overstate or underestimate the support for various candidates. There are a lot of factors which can determine a poll’s accuracy, with the phrasing of the questions being a key component.

Forty years ago this month, a young assistant psychology professor from Wayne State University published the results of a study in the February 1968 edition of the journal Psychological Reports:

That a poll can be biased by the statement of the question and/or the attitudes of the interviewers is part of the pollster’s lore. The degree of biasing possible is largely unknown and the present study was undertaken to aid in the anchoring of this parameter.

Twenty introductory psychology students interviewed by telephone 590 fellow-students randomly selected from the student directory while an additional 574 were personally interviewed utilizing quota sampling. In either case, the interviewer asked [the subject] two questions from an array of 18, one concerning Johnson’s policy in Viet Nam and the other about its news coverage. No [subject] was asked two questions of the same coloration (neutral, agree-, or disagree-biased). Interviewers were instructed to promote agreement with the casting of each item by acting as though they personally endorsed it by tone of voice (and, in the face-to-face interview, facial expression).

And what did this young neophyte professor discover? Depending on how the question was phrased and how it was asked by the interviewer, it was possible to manipulate a change in opinion of around 13%. That young researcher concluded:

Considering the influence of an interviewer in a transitory situation, social psychologists would do well to cast a critical eye on classical assessment-message-assessment attitude experiments.

That young assistant professor was Paul Cameron, who would later become famous for being a prolific generator of junk statistics for the anti-gay industry. Little did he know how useful that little two-page study would become later in his career.

Source: Cameron, Paul; Anderson, James. “Effects of introductory phrases and tonal-facial suggestions upon question-elected responses.” Psychological Reports 22, no. 1 (February 1968): 233-234.

Comments

POST COMMENT | COMMENT RSS 2.0 | TRACKBACK URL

Bruce Garrett
February 11th, 2008 | LINK

Good Catch!

David
February 11th, 2008 | LINK

Very interesting.

1968 was 40 years ago, not 30.

Jim Burroway
February 12th, 2008 | LINK

That’s what happens when you’re in a hurry! Thanks!

Leave A Comment

All comments reflect the opinions of commenters only. They are not necessarily those of anyone associated with Box Turtle Bulletin. Comments are subject to our Comments Policy.

(Required)
(Required, never shared)

PLEASE NOTE: All comments are subject to our Comments Policy.