APPENDIX EE. The variable "value of college" was created by J. Michael Armer for his dissertation "Community and School Environments and College Plans of Public High School Seniors", University of Wisconsin, 1964. The passage on the following pages is from this dissertation. Measures of both college-going climate conceptions have been developed for the present study. The two school climate indexes are (1) the percentage of the senior class planning on going to college, and (2) the average value of college as perceived by the seniors of each school. The first emphasizes modal college-going plans in the schools, and the second, normative climates or the value of college-going at each school. School climates alone will be measured since comparable data were not available for communities. A further word of explanation is in order regarding the measurement of the two school climate variables. The first index, of course, is a simple percentage based on the individual college plans of all the students from each school. The second index of school climate is an average of the scores of the seniors in each school in response to a series of eighteen weighted "value statements" regarding their perception of the "value of going to college."^1 The value statements are presented below (with the respective positive and negative weights added adjacent to each statement): Weights -70 I would rather start earning money quickly, and learn on the job. 38 I (am) would be greatly dissatisfied to stop at my present level of knowledge. 10 College life and activities (like athletics) attract me very much. 27 College graduates get jobs with better pay, 27 The country needs more people who have highly developed skills and knowledge. 5 College is a good place to meet a worthy life-mate. -42 Skilled laborers get paid as such as most college graduates. 23 Going to college enables you to study more lines of work before deciding on a career. 33 A college education helps you live a happier, more complete life. -24 Going to college costs more than it is worth. 22 College studies will make you work at a high intellectual level, and I like that. 1): Questionnaire item 30. See Appendix G. 16 College graduates usually have the leadership positions. -55 Learning on a job is more practical than most school learning. -7 Persons who do not have college educations often make better leaders. 37 College life broadens you socially, and develops your personality. -41 Success in life depends upon ability and effort, not amount of education. -47 Going to college would be a waste of time for me. 11 Going to college has just be accepted; I have never thought of anything else. The weights for the value statements were derived by factor analysis using the principal components method.^1 Loadings of the eighteen statements on the principal-axis of the unrotated factor matrix are defined as the relative positive or negative weights of each college going value statement. The primary assumption involved in the weighting system is that the principal-axis is the "value of going to college" factor or dimension which the statements were designed to measure.^2 Each student's responses to the statements were weighted using the obtained factor loadings and averaged for a "value of college scare. The value scores for the students were then averaged for each high school to obtain the "average value of college" as an index of the climate of each school. The scores for the percentage of seniors in each school planning to go to college as perceived by the seniors of each school were added to the other information available on each student for use in correlation analysis of the community and school status thesis. 1.): See Benjamin Fruchter, Introduction to Factor Analysis (New York: D. Van Nostrand Company, Inc., 1954), pp. 99-100 for a discussion of the principal-axes method. A discussion of fundamental ideas and terms can be found on pages 1-50. More general issues are in Raymond B. Cattell, Factor Analysis: An Introduction and Manual for the Psychologist and Social Scientist New York: Harper and Brothers, 1952). 2.): The validity of this assumption may be intuitively checked by comparing one's personal evaluation of the pro or con direction of the eighteen value statements with tke positive and negative loadings given by factor analysis. College Plans: The dependent variable, college vs. non-college plans is provided by students' responses to two items asking them of their post-high school plans.^1 The first item inquires about the general plans of students after high school--to get a job, continue school, enter military service, get married, and so forth. The second item asks students to indicate the kind of schooling contemplated if they stated that they plan to continue their formal education after high school. The various alternatives to this question have been dichotomized for purposes of this study into (1) plans to attend college or university upon graduating from high school, and (2) does not plan to attend college or university after graduation. In other words, respondent has the option to terminate formal education, get a job, and so forth, or to continue with any kind of schooling other than college or university. Of the 3999 Milwaukee metropolitan area seniors, 1460 or 36.5 percent indicated plans to continue to college or university after graduating. MEASUREMENT OF THE "AVERAGE VALUE OF GOING TO COLLEGE" WITHIN HIGH SCHOOLS In order to devise an index of the status of going to college in different high schools, information was first required on individual student perceptions of the value of going to college. The original survey obtained information on the following item: 30. Which of the following statements best describe your opinion of the value of going to college? (Answer whether you plan to go to college or not. Check the 3 statements which seem most important to you.) An unfortunate aspect of this question was that the student was instructed to check only three statements. Thus 15 of the 18 statements are not checked by each student. This means that each statement had a theoretical probability of 17.7 percent of being checked. The actual frequency ranged from a low of 3 per cent who checked item 10, to a high of 34 per cent who checked item 5. Therefore, when the statements are scored zero (0) or one (1) for omission or selection respectively, it is not surprising that correlations between the statements rarely exceed r = + or - 30. However, the range of correlations from negative to positive suggests that some statements "go together" better than others, while other statements do not cluster. It seems highly plausible that students who check a statement at a given value level tend to check other statements at similar value levels, even though various other less pervasive factors also undoubtedly influence the selections. The stronger positive and negative correlations should reflect statements more nearly equal and greatly opposed to each other in the degree college-going is valued. 1.): See (Appendix G,) Questionnaire item 1 and 2. A useful statistical technique for systematically developing notions concerning the main dimensions underlying a domain of value, attitudes, behavior or other classes of variables is factor analysis. Therefore the data were factor analyzed using the principal components method in order to answer the question whether "value of going to college" can reasonably be identified as the principal dimension measured by the set of 18 value statements. In the principal-axes method of factoring, each factor extracts the maximum amount of variance, leaving the smallest possible residuals and condensing the correlation matrix into the smallest number of independent factors. (A full account of the method can be found in Fruchter's Introduction to Factor Analysis 1954 or Cattell's Factor Analysis for Social Sciences 1952). The following principal-axis was extracted with items rearranged according to loadings (No other factor was extracted with eigenvalues exceeding unity.): Loadings on Principle- Axis Value Statements -70 1. I would rather start earning money quickly, and learn on the job. -55 13. Learning on the job is more practical than most school learning. -47 17. Going to college would be a waste of time for me. -42 7. Skilled laborers get paid as much as most college graduates. -41 16. Success in life depends on ability and effort, not amount of education, -24 10. Going to college costs more than it is worth. -7 14. Persons who do not have college educations often make better leaders. +15 6. College is a good place to meet a worthy life-mate. +10 3. College life and activities (like athletics) attract me very much. +11 18. Going to college has just been accepted; I have never thought of anything else, +16 12. College graduates usually have the leadership positions. +22 ll. College studies will make you work at a high intellectual level, and I like that. +23 8. Going to college enables you to study more lines of work before deciding on a career. +27 5. The country needs more people who have highly developed skills and knowledge. +27 4. College graduates get jobs with better pay. +33 9. A college education belps you live a happier, more complete life. +37 15. College life broadens you socially and develops your personality. +38 2. I (am) would be greatly dissatisfied to stop at my present level of knowledge. It will be observed that the scoring system of having students select only three statements not only contributes to low inter-statement correlations, but also to low factor loadings since these are based on the interstatement correlations. The fact that the variance in the 18 statements explained by the principal factor is only 2.1, or approximately 11 per cent of the total variance, reflects the method of scoring only 3 of the 18 items and suggests the existence of other minor dimensions measured in the data. Examination of the statements and the corresponding loadings suggests that the principal axis may be identified as a "value of going to college" dimension and, thus, the loadings were accepted as value weights of each statement for developing college-going value scores for students and schools. The value of going to college as perceived by each student was defined as the average of the positive and negative loadings on the several value statements checked. Since some students mistakenly selected only one statement, the theoretically possible range of value scores was -.70 for a very negative evaluation of the value of going to college, to +.38 for a high evaluation of college. A transformation of the scores according to the following formula gave a new theoretical range of 1 to 98 which is compatible with ranges on other variables in the analysis and more convenient for data processing purposes: Value score = (Average loading x 90) + 65. The value of going to college was then averaged across students within each high school to give the "average value of college" for each school, that is, the school college-going climate variable used in the analysis. As a further check on the validity of the "average value of college" variable.. it was correlated with the "per cent of seniors going to college," and found highly correlated; r = .85. DISTRIBUTION OF COMMUNITY AND SCHOOL STATUS AND CLIMATE VARIABLES Occupational Status* Communities Schools Score N Score N 13 126 54 152 21 241 55 179 23 228 57 198 24 105 58 153 29 288 59 384 30 279 60 495 31 438 61 587 32 358 62 674 33 153 63 438 34 152 64 105 47 390 65 232 41 105 68 251 43 388 72 151 54 75 69 346 72 176 83 151 Mean = 38.8 61.4 S.D. = 17.1 3.8 Educational Status** Communities Schools Score N Score N 6 367 5 290 10 1117 6 312 11 458 7 743 13 274 9 799 14 152 10 227 15 207 11 384 16 388 18 571 20 183 41 176 28 75 62 151 30 346 43 176 57 151 Mean = 16.9 15.0 S.D. = 11.8 13.2 School College-Going Climate*** % Going Average Value Score N Score N 6 49 60 726 20 394 61 75 21 305 63 258 23 288 64 207 26 207 65 465 27 75 67 284 28 543 68 412 33 469 70 279 34 152 71 49 35 177 72 183 37 279 74 346 57 346 75 176 59 388 76 388 61 176 86 151 82 151 Mean = 36.7 68.1 S.D. = 16.7 6.4 *Community occupational status = percent of all employed males, 14 years and older who are employed in white-collar occupations. *School occupational status = percent of all seniors' fathers employed in white-collar occupations. **Community educational status = percent of all community adults 21 years and older with one year or more of college education. School educational status = percent of all seniors' parents with one year or more of college. ***Modal college plans = per cent of senior class in each school planning on going to college. Average value of college = average of the "value of college" scores of seniors for each school.