Suchergebnisse

Though interviewer ratings are used in assessing the SES of R homes in survey res, they are not very reliable, & fail to differentiate between households especially in the middle group. The scale provides valid & reliable SE groupings of people interviewed in the street. The requirements of a worthwhile scale were that it (a) be reliable & valid, (b) differentiate between homes all along the dimension of SES, (c) be simple to use in the field, (d) be readily processed, & (e) not require frequent revision. 7 questions: (1) Who is the chief wage earner? (2) What is wage earner's occup, (3) educ of wage earner, (4) automobile ownership, (5) telephone `ownership', (6) home ownership, & (7) number of bedrooms & number of people living in dwelling, were used in a national personal interview survey of 7,500 Ur & Ru homes in Feb 1951. 5 of these were scaled against interviewer ratings of SES. The sample was from 273 cities, towns & villages within which interviewers were assigned to spots on the basis of SE maps of the places. Each interviewer had a quota of upper, middle, & lower group interviews. Interviews were held with only one person per household & R's were chosen to match the distribution of individuals aged 8+ yrs in the US pop as shown by the Census of 1950. The criterion for scaling the answers to the questions was interviewer ratings. A detailed analysis of the scaling procedure for one question is presented. The scale discriminates & the distributions approximate the normal curve. A cross-validational study produced a Pearsonian r=.61 between interviewer ratings & (total - sum) scaled scores which is as high as the r obtained between ratings themselves when diff interviewers were used to rate the same households. J. D. Twight.