For political scientists and pollsters the way the individual voted on previous occasions provides an important source of data. In the absence of longitudinal studies, recall of past vote tends to be taken as equivalent to actual vote cast. How accurate is such recall? How far does accuracy decrease with time, where recall concerns not one, but two, previous elections? How far do errors introduce a systematic bias in the conclusions drawn from such data?

In England in 1951, as part of a larger inquiry, 13–14‐year‐old boys from middle and working class homes were systematically assessed as to the role of media in their lives. In 1962, when they were 24–25, 365 of the group completed a questionnaire about leisure and the media; and in 1970, 246 of the sample, then aged 32–33 were again questionned. The relative role of social, personality, and outlook factors in accounting for media usage and taste was examined at the several developmental stages, as well as longitudinally across the 20 year span of the study, which encompassed the period during which television was introduced and absorbed into the leisure space of the sampled population.

No abstract

No abstract

No abstract