A principle-factor analysis of the 88 items comprising the Survey of Reading Attitudes investigated the construct validity of the eight dimensions of reading attitude. The sample consisted of 344 inner-city Detroit fifth graders. The factor structure showed support for the Expressed Reading Difficulty, Reading as Enjoyment, Silent vs. Oral Reading, and Alternative Learning Modes dimensions. However, the Reading Anxiety dimension tended to merge with the Expressed Reading Difficulty dimension, and the Reading a s Direct Reinforcement and Reading Group dimensions merged into a single dimension. This finding suggests these dimensions may be correlated, and an oblique rotation may be more appropriate. The Comics dimension split into two dimensions: Comic Books and Newspaper Comics. Thus, these findings generally cross-validate the dimensions of reading attitude found in earlier studies of the Survey of Reading Attitudes.
Recent studies by Engin, Wallbrown, and Brown (1976) and by Wallbrown, Brown, and Engin (1978) have demonstrated that reading attitude is a multifactored phenomenon. The most recent edition of A Survey of Reading Attitudes by Wallbrown, Brown, and Engin (Note 1) was shown to have eight orthogonal dimensions of reading attitude for intermediate-grade children (Wallbrown, Brown, & Engin, 1978). The eight attitude dimensions were Expressed Reading Difficulty, Reading as Direct Reinforcement, Reading as Enjoyment, Alternative Learning Modes, Reading Group, Reading Anxiety, Silent vs. Oral Reading, and Comics.
The present study seeks to cross-validate the dimensions of reading attitude found in the earlier studies. Since Brown, Engin, and Wallbrown (1979) found developmental changes in reading attitudes from grades four to six, the current study concentrates on fifth graders. The sample from inner-city Detroit differs substantially from those of a mid-sized central Ohio city used in the previous studies.
METHOD
Sample
The sample consisted of 344 inner-city children from four elementary schools in the Public School System of Detroit, Michigan. All children enrolled in regular fifth-grade classrooms and present on the testing day were included. This included 173 males and 171 females, with 90 white and 254 black children, The mean age was 11.3 years, with a standard deviation of .8 years.
Procedure
The most recent edition of A Survey of Reading Attitudes (Wallbrown, Brown, & Engin, Note 1) was administered to classroom-size groups. This edition contained 88 items designed to measure eight different reading attitude dimensions. Because of the possible presence of pupils with serious reading problems, the items and directions were included in a taped presentation that accompanied the test booklet. Children expressed