Lost and Found, Letters and Methods: Assessing Attitudes toward Chiropractic and Medical Care

• Main Authors: Hannah Kern, William H. Yeaton
• Format: Article
• Language: English
• Published: University of Arizona Libraries 2010-10-01
• Series: Journal of Methods and Measurement in the Social Sciences
• Online Access: https://journals.uair.arizona.edu/index.php/jmmss/article/view/78

Attitudes toward traditional and chiropractic medicine were compared using Milgram's lost letter technique. A total of 192 letters were placed on the windshields of vehicles in parking lots at six restaurants and department stores in each of four quadrants of a medium-sized, Southeastern city. These letters were addressed to "Admissions" at either a fictitious Institute of Medicine or Institute of Chiropractic Care. Return addresses included either a male or a female name. Thus, those who found a lost letter were faced with the option of returning or not returning a letter from either a male or a female, addressed to an Institute of traditional or non-traditional medicine. After examining previous studies which had used the lost letter technique, numerous methodological improvements were implemented. For example, letters were randomly assigned to potential drop spots for each of 24 study locations (six study locations in each of four city quadrants), and a Latin square design was used to control for possible order effects in the four study conditions that were implemented. Nearly 65% of the letters (124 of 192) were returned. We found: 1) letters addressed to a fictitious Institute of Chiropractic Care were just as likely to be returned as those addressed to a fictitious Institute of Medicine; 2) letters with female return addresses were as likely to be returned as those with male return addresses; 3) there was no interaction between study conditions; 4) based on what was essentially a replication study, a comparison of the pattern of returns using the first and second cycle of lost letters (n = 96 for each cycle) revealed an equivalent pattern of no-difference findings.   DOI:10.2458/azu_jmmss_v1i1_kern