The statements in this scale were first published by Thurstone (1932) and were reprinted by Shaw and Wright (1967). Beside each statement is the Thurstone scale value published with the original scale, followed by the new Thurstone scale value and associated standard deviation calculated by Roberts (1995). The new scale values were developed using a variant of the successive intervals procedure which allowed scale value standard deviations to differ across items (Torgerson, 1958). Scale values were based on favorability judgments from 90 college students who rated each statement on a 9-point scale. The endpoints of the rating scale were anchored with the terms "very unfavorable" and "very favorable", respectively. Eight of the original items could not be adequately scaled due to their ambiguity or their failure to fit the Thurstone model.
In addition to scaling statements, Roberts (1995) had 245 subjects indicate the extent to which they agreed with each of the 24 statements. Responses were on a 6-point rating scale where 1=Strongly Disagree, 2=Disagree, 3=Slightly Disagree, 4=Slightly Agree, 5=Agree and 6=Strongly Agree. The data are formatted as follows:
| Field | Columns |
| Subject ID | 1-4 |
| Item 1 | 6 |
| Item 2 | 7 |
| Item 3 | 8 |
| . | . |
| . | . |
| . | . |
| Item 24 | 29 |
Further information about the scale values and the graded disagree-agree data can be found in Roberts (1995).
Roberts, J. S. (1995). Item response theory approaches to attitude measurement. (Doctoral dissertation, University of South Carolina, Columbia, 1995). Dissertation Abstracts International, 56, 7089B.
Shaw, M. E., & Wright, J. M. (1967). Scales for the measurement of attitudes. NY: McGraw-Hill.
Thurstone, L. L. (1932). Motion pictures and the attitudes of children. Chicago: University of Chicago Press.
Torgerson, W. S. (1958). Theory and methods of scaling. NY: Wiley.