The validity of the Southam News 1987 survey of literacy in Canada was questioned relative to its content and structure. The survey viewed literacy as a set of complex information processing skills and adopted a hierarchical view of these skills. This hierarchy model set up illiteracy as a deficiency. The Southam study's underlying assumption was that literacy was a necessary condition for functioning in Canadian society. To assess the validity of the survey, both the questionnaire content and the survey environment were examined. Content validity was assessed by looking at the questionnaire tasks and deciding whether they were important for functioning in Canadian society. Findings showed the inability to answer the questions correctly did not reflect the inability to get by in life because no allowance was made for solving the problems in other ways. Construct validity was judged by determining whether the questionnaire measured ability to function in life rather than ability to do something else. Cases showed that the survey tasks measured reading ability, not ability to get by in life. The study assumed unwisely that the survey had predictive validity--that if someone could answer a survey question, he or she would be able to cope with the same problem in life. The validity of the survey could also have been affected by other factors: time constraints placed on participants; use of only printed materials; print size; isolation; and interpretation of responses. (YLB)