Search

Discusses the use of computer administered testing by The American College. Student performance on computer administered versus paper-and-pencil tests is examined, student attitudes about exams are described, the effects of time limits on computerized testing are considered, and offline versus online testing is discussed. (31 references) (LRW)

Background. Data suggest that colorectal cancer could be cut by approximately 60% if all people aged 50 years or older received regular screening. Studies have identified socio-cultural attitudes that might inform cancer education and screening promotion campaigns. This article applies item response theory (IRT) to a set of survey items selected to assess sociocultural attitudes in order to determine how current measures may affect what we know about how these attitudes affect colorectal cancer screening (CRCS). Design and Methods. A survey of colorectal cancer screening, screening attitudes and cultural beliefs was administered to 1021 African Americans - 683 women and 338 men, ages 50 to 75. Eligibility criteria for participation included being born in the United States, selfidentified African American male or female, age 50 to 75 years. The IRT analysis was performed on 655 individuals with complete data for the 43 observed variables. Results. Twenty-nine items comprise the Multi-construct African American Cultural Survey (MAACS) that addresses seven cultural constructs: mistrust/distrust, privacy, ethnic identity, collectivism, empowerment, and male gender roles. The items provide adequate information about the attitudes of the population across most levels of the constructs assessed. Among the sociocultural variables considered, empowerment (OR=1.078; 95% CI: 1.008, 1.151) had the strongest association with CRCS adherence and privacy showed promise. Conclusions. The MAACS provides a fixed length questionnaire to assess African American CRCS attitudes, two new constructs that might assist in CRCS promotion, and a suggested focus for identification of additional constructs of interest. [ABSTRACT FROM AUTHOR]