1 Paper given at the 1990 Conference of the Australian Association for Research in Education Favourable attitudes towards school and students' perceptions of curriculum choice. Michael Sheret Linda Fink John Ainley The Australian Council for Educational Research _____________________________________ Introduction This paper describes part of the research of a project commissioned in 1987 by the New South Wales Department of Education. The project is still in progress. It has a strong longitudinal component with a large cohort of students (originally more than 3000 from 22 government high schools) being surveyed from Year 9 to Year 12 of their school years and beyond. In addition to the longitudinal cohort, viz. Year 9 students in 1987, students from Years 10 and 11 were also surveyed in 1987 in order to obtain cross-sectional base data. The project was conceived during a period of increasing retention rates into the post compulsory years of schooling. The overall aim of the project is to examine the factors which influence students to stay on at school and gain from the experience in terms of both attainment and broader definitions of a satisfying school experience. The more quantitative parts of the present paper deal with data from some 1600 Year 11 students surveyed in 1987 and data from some 2000 Year 11 students in 1989 (different students, same schools). The paper examines Year 11 students' attitudes in several dimensions to their school experience, and how these attitudes are influenced by the background characteristics of the students and by the students' own perceptions of the degree of curriculum choice within the school. Then the paper looks at some school and student characteristics to see if these can explain why some students are more satisfied than others with the subjects available for their curriculum choice. With the two data sets the paper is able to examine whether patterns observed in 1987 are sustained in 1989. The Year 11 students in 1989 belong to the main longitudinal cohort; consequently there are more data on these students, and this allows a more extensive analysis for these students. We were inspired to look closely at our quantitative data and investigate the links between students' perceptions of curriculum choice and their favourable attitudes to school, as a result of some qualitative data gathered in Term 3 1988. The longitudinal cohort of students was then in Year 10 and was surveyed by means of the usual quantitative questionnaires. In addition we selected at random 15 students from each school, 330 students in all, and gave them a group interview. Among the questions asked was: "If you had the authority to do so, name one change you would make to your school?" We made sure that every student contributed one suggestion for the change they considered the most important. A suggestion could be unique or shared with another student, provided the student considered their suggestion to represent the most important change. After the round of interviews we categorised the 330 suggestions for change that we had gathered. Concern with the curriculum, expressed mainly in terms of subject choice both for the Years to Year 10 and the impending Years 11 and 12, accounted for 31.2% of the students' responses. The other major categories were improvements in physical facilities (17.3%), changes in the schools' behaviour rules (14.5%) and improvements in teachers' performance and behaviour (14.5%). Other qualitative indications of students' concern with curriculum choice occurred in 1990, our final year of data gathering. We gave group interviews to 50 students from five selected schools in Term 4. The students were Year 11, not part of our main cohort. The schools were selected because they exhibited unique characteristics, such as high achievement and high satisfaction with school life, and became the subject of three-day qualitative observation visits. The student group interviews were largely unstructured. Concerns over restrictions on subject choice were present in all five schools. Students were aware of the reasons for these restrictions - school size, low enrolment for a particular course, availability of teachers - but this did not appear to diminish their sense of missing out. Most concern was expressed over Board determined courses, for example some 3-unit options not offered or two courses in a timetable clash. Some concern was expressed with other courses, for example the failure of their school to present performing arts courses. Other qualitative indicators of student concern over the curriculum came from reading through the completely unstructured section of our "School and You" written questionnaire given to Year 12 students (the main cohort) in Term 2 1990. Here we let the students speak for themselves. "I feel for senior students, there should be a wider range of subjects to choose from and no restrictions as to what you can study". "You should be able to pick the courses you want and not what is available on the lines". "Single sex school with limited subjects". "The subject choices aren't fantastic but I have learned to live with them. I feel that computer courses should be catered for better with the advancement of technology". "More subjects should be offered that relate to the performing arts. It is ridiculous that a creative person should have to do science oriented subjects because there are no other choices". In the following sections we shall proceed with the analyses of our quantitative data. Students' Perceptions of the Quality of their School Life For this paper students' perceptions of the quality of their school life (QSL) are measured using the Australian Council for Educational Research (ACER) 40- item Quality of School Life questionnaire. This instrument has been widely used. It has a robust factorial structure (Williams and Batten, 1981), which has been confirmed by routine analysis for this data set. It measures students' satisfaction with their school experience on seven subscales. These subscales are:-POS: general positive feeling towards school, TCH: satisfaction with teacher-student interaction, STA: feeling of status accorded to student by significant others within school, SOC: sense of learning about and getting along with other people, OPP: belief in the relevance of schooling, ACH: confidence in one's ability to be successful at school work, NEG: negative feelings, anxiety related and not a polarisation of POS. The instrument does not envisage that these seven domains are independent of each other but that together they constitute a view of the elements of school life for students. Accordingly the subscales are intercorrelated; the median value of the correlation between subscales is 0.40. The QSL instrument is reproduced in Appendix A. Table 1 show some properties of the QSL subscales. Table 1 Properties of the Quality of School Life Scale Subscale Alpha Mean Standard Possible Reliability Deviation Range POS .80 (.82) 13.7 (14.0) 2.7 (2.6) 5-20 TCH .82 (.84) 16.6 (17.3) 3.0 (2.9) 6-24 Main figures refer to 1987 STA .77 (.81) 16.1 (16.5) 2.7 (2.7) 6-24 data set. Figures in parentheses SOC .77 (.78) 18.4 (18.6) 2.5 (2.4) 6-24 refer to 1989 data set. OPP .85 (.84) 18.1 (18.3) 3.2 (3.1) 6-24 ACH .80 (.81) 17.6 (17.8) 2.6 (2.6) 6-24 NEG .77 (.80) 10.0 (10.0) 2.5 (2.7) 5-20 _______________________________________________________ Students' Perceptions of Curriculum Choice If students are able to take any combination of subjects from a very wide range of subjects and levels of difficulty, it is expected that they will be happy with their choice or at least they will not feel that curriculum choice has been thrust upon them. In practice schools, mainly through circumstances but sometimes partly as a result of timetabling inflexibility, offer a limited range of subjects and a restricted choice within that range. Consequently students have positive feelings about some subjects as representing their true choice and negative feelings about other subjects which they feel they have been compelled to do for various reasons, viz. limited menu of subjects, prerequisites and compulsory subjects. The relevant question asked of students is reproduced in Appendix B. Students listed the subjects they were studying in Year 11. Next to each subject they ticked the one of two boxes which better represented their feelings about that subject: either "I am doing this subject because it is something I want to do, or it is my choice, or I think it will be useful for me" or "I am only doing this subject because it is compulsory, or I was told I ought to do it, or it was needed before I could do other subjects, or it was the one available on the timetable". The precise meaning of these alternatives is less important than their capacity to act as appropriate triggers. Most students listed six or seven subjects. The percentage of positive responses was calculated as the variable CUR and used to indicate the extent to which the student felt that they had a choice in their program of studies. The question asked of students in 1989 was slightly different from that presented in Appendix B. Each listed subject in 1989 was accompanied by the number of units for that subject (1 or 2 or 3 or 4). The number of units is proportional to the amount of contact teaching time for that subject and the weight given to the subject in computing the Tertiary Entrance Score, and it is an indication of the level to which the subject is studied. When the subject is weighted by the teaching time devoted to it, then if necessary a modified and more refined version of CUR can be calculated, viz. the percentage of units eliciting a positive response. Student Background Characteristics Four background characteristics of students are used in the regression analyses as independent variables. The family socioeconomic background variable, SES, is on a six-point ordinal scale. Data were gathered on both father's and mother's occupational status using a sixteen point scale devised by the Australian National University (Broom et al, 1977). Many responses on mother's occupation were difficult to classify. Therefore, only the father's occupation was used and collapsed into six categories as follows: unskilled 1, semi-skilled 2, skilled 3, clerical & service 4, managerial 5, professional 6. Collapsing of categories loses some detail but results in fewer erroneous classifications and fewer near-empty categories. The extent to which a student comes from a non-English speaking background, NES, is on a four-point ordinal scale. Data were gathered in considerable detail about student's, mother's and father's country of birth. Collapsing of data was necessary to avoid near-empty cells. In this collapsed classification a very small number of students may be illogically assigned and no distinction is made among English speaking countries and among non-English speaking countries. The categories are as follows: student and both parents born in ES country 1, one parent born in NES country 2, both parents born in NES country 3, student born in NES country 4. The highest numbered category supercedes any lower numbered classification. Parents differ in the extent to which they encourage their children to stay at school and/or undertake further study. A variable, PAR, is computed from a question (not originally designed for this purpose) asking each student to categorise their parent's plans for the student's future on leaving school. PAR is a four-category ordinal variable according to the parent's plans or expectations for their child: full-time job 1, part-time job 2, don't know 3, part-time or full-time study 4. The highest numbered category supercedes any lower numbered classification. Students who gave an open response of the kind, "My parents want me to do whatever I want to do", were assigned a value corresponding to their own plans as specified in an earlier question. The "don't know" category is felt to be less discouraging of post-school education than a clear expectation that the student will enter the workforce immediately on leaving school; this is perhaps not entirely logical; however, subsequent checks reveal that the chosen progression of values relates monotonically to the dependent variables under consideration. The variable GEN denotes the gender of the student : 1 for male and 2 for female. When GEN is used as an independent variable in a regression analysis it measures the influence that being female has on the dependent variable. The are no problems of multicollinearity among these four background variables. Intuition might lead one to suppose that PAR and SES correlate closely, but this is not the case, their correlation is only 0.10 for the 1987 data set and 0.12 for the 1989 data set. Influences on Perceptions of the Quality of School Life There are fourteen regression analyses shown in Table 2, seven for 1987 data and seven for 1989 data. Each analysis takes one of the seven subscale scores from the quality of school life question as the dependent variable and examines the effect of students' perceptions of curriculum choice after controlling for student background characteristics. Table 2 Influences on QSL subscales Dependent variables (QSL subscales) ____________________________________________________________________ Indep POS TCH STA SOC OPP ACH NEG Variables PAR .12* (.13*) .14* (.12*) .03 (.04) -.03 (- .03) .14* (.10*) .14* (.15*) -.05 (.03) SES .01 (-.05*) .01 (-.01) .03 (-.04) -.01 (- .02) .00 (-.04) .04 (-.05*) .04 (.05) GEN .04 (.07*) .05 (.06*) -.04 (.02) .05 (.14*) .00 (- .02) -.01 (-.02) .09 (.05) NES .08* (.08*) -.09* (-.05) -.03 (- .01) .01 (.01) .07* (.04) .04 (.02) .04 (.06*) CUR .27* (.22*) .21* (.19*) .17* (.18*) .10* (.11*) .24* (.28*) .20* (.21*) -.18* (-.15*) Mult R2(%) (B'ground 11.1 (9.0) 7.3 (5.6) 3.2 (3.5) 1.4 (3.2) 9.7 (9.6) 7.4 (7.7) 4.2 (3.1) & CUR) __________________________________________________________________________________ Mult R2(%) (B'ground) 4.0 (4.0) 3.2 (2.1) 0.5 (0.4) 0.3 (1.9) 3.8 (1.9) 3.3 (3.1) 1.1 (0.7) Mult R2(%) (CUR) 8.2 (5.6) 4.7 (3.8) 2.8 (3.2) 1.1 (1.3) 6.9 (8.1) 4.9 (5.1) 3.2 (2.2) __________________________________________________________________________________ The first five lines of the table show standardised regression coefficients; those marked with an asterisk are significant at the p < .01 level. Main figures refer to the 1987 data set; figures in parentheses refer to 1989 data set. For simplicity it is best to refer to the 1987 and the 1989 data sets separately. For the 1987 data the regression coefficients in the first five lines of Table 2 show that for each of the seven QSL subscales CUR, the student's perceptions of curriculum choice, is a much better predictor of a student's having favourable attitudes to school than any of the student's background characteristics. Comparing the R2 (Background & CUR) to the R2 (Background only) shows that by putting CUR in the regression equation the amount of variance explained increases substantially. For example, for POS (a general positive feeling towards school) CUR raises the variance explained by the four background variables alone from 4.0% to 11.2%. It is also interesting to note that CUR and the group of four background variables do not strongly correlate with each other. For example, for the POS subscale, R2 (Background only) plus R2 (CUR only) equals 4.0 + 8.2 which is only slightly greater than 11.2 the value of R2 (Background & CUR). For the 1989 data, which are shown in parentheses in Table 2, the patterns of 1987 are confirmed with one exception. In the regression where SOC (sense of learning about and getting along with other people) is the dependent variable multiple R2 increased in 1989 with both CUR and the four background variables in the regression. This has come about because of the variable GEN; in other words being female was positively associated in 1989 with having a high sense of social satisfaction with school. Students' Perceptions of Curriculum Choice and School Membership Are students' perceptions of curriculum choice influenced by their being a member of one school rather than another? The intuitive answer would be "Yes". Table 3 presents some data which can put this to the test. For the 1987 data there is variance among school mean values for CUR with values ranging from 64.3 for school U to 81.6 for school S. This variance is significant at the p < .05 level but it is much less than the total variance among individual students (standard deviation = 17.8). ETA2 =2.3% indicating that school variance explains 2.3% of the total variance. School U stands out from the others in having a very low value of CUR; school U is a very small school with the lowest number of Board determined courses on offer. Turning to 1989 data the variance among schools remains much the same. The school mean values for CUR range from 72.1 for school H to 83.4 for school D. This variance is significant at the p < .05 level but it is much less than the total variance among individual students (standard deviation = 19.0). ETA2 =2.6% indicating that school variance explains 2.6% of the total variance. School U no longer stands out from the others; the value 75.7 of CUR for school U is close to the average. There is no obvious reason for this but it should be noted that schools low on satisfaction with curriculum choice in 1987 do not necessarily retain this characteristic into 1989. While the correlation between 1987 and 1989 values of CUR is positive, it is not a strong correlation, 0.18, which, with only 22 cases at school level, is not statistically significant. Table 3 School Aggregated Data on Satisfaction with Curriculum Choice and Data on Enrolments and Curriculum Offerings School Mean CUR SD CUR N OAS Board Enrolment A 78.9 (74.0) 17.8 (24.1) 42 (30) 7.4 (0.1) 26 (19) a B 73.9 (73.7) 17.4 (18.9) 94 (114) 3.2 (1.9) 31 (37) b C 74.4 (77.9) 19.0 (17.9) 62 (103) 5.6 (3.8) 27 (31) d D 80.1 (83.4) 13.9 (16.2) 84 (126) 7.7 (5.3) 36 (37) d E 77.3 (79.1) 17.0 (18.0) 59 (106) 5.4 (7.9) 26 (34) d F 75.5 (73.8) 16.9 (21.2) 99 (115) 5.6 (8.4) 30 (32) b G 75.1 (79.1) 19.9 (19.0) 89 (111) 6.2 (5.6) 27 (28) g H 78.9 (72.1) 13.9 (23.3) 89 (101) 5.8 (6.4) 36 (31) d I 76.4 (79.9) 18.4 (19.6) 114 (116) 6.8 (6.3) 29 (50*) g J 76.3 (78.4) 20.9 (19.9) 50 (68) 5.9 (8.5) 30 (32) g K 74.5 (80.0) 19.8 (16.8) 59 (91) 10.8 (7.1) 29 (30) g L 77.8 (77.9) 18.8 (17.9) 100 (87) 6.0 (3.4) 35 (31) d M 74.9 (78.5) 20.8 (21.3) 48 (137) 4.0 (6.4) 28 (30) g N 78.0 (81.7) 17.5 (16.2) 85 (111) 7.3 (7.5) 29 (30) g O 76.8 (75.8) 18.0 (19.9) 81 (124) 4.3 (6.0) 28 (29) b P 74.5 (81.1) 15.3 (15.8) 81 (96) 2.3 (3.8) 31 (35) b Q 74.0 (80.7) 16.9 (19.6) 20 (26) 5.7 (4.2) 23 (24) a R 75.4 (81.2) 17.2 (17.6) 111 (112) 14.2 (9.1) 31 (32) g S 81.6 (81.6) 18.6 (16.8) 103 (98) 9.9 (7.1) 28 (30) g T 80.3 (75.7) 17.3 (19.5) 44 (84) 6.9 (8.9) 26 (32) b U 64.3 (75.7) 19.9 (14.8) 25 (23) 6.5 (10.3) 22 (18) a V 79.3 (80.7) 17.6 (17.1) 56 (67) 0.0 (6.6) 24 (29) b All Schools 76.7 (78.4) 17.8 (19.0) 1595 (2046) ___________________________________________________________________________ Main figures refer to the 1987 data set; figures in parentheses refer to the 1989 data set. OAS = estimate of course enrolments in Other Approved Studies courses as % of total course enrolments. Board = number of Board determined courses on offer. Enrolment = total school enrolment (mean for July 1987 and July 1989). a œ 350. 351 œ b œ 825. 826 œ g œ 950. d ò 951. * The unusually high number, 50, of Board determined courses in 1989 for school I is due to the variety of low enrolment language courses arranged by the school with external agencies. Are schools where the mean level of satisfaction with curriculum choice is higher generally the schools offering a greater variety of courses? The expected answer is "Yes". To test this hypothesis a regression is used where CUR is the dependent variable with OAS and Board as two independent variables. In 1987 the multiple R is 0.38 and the beta weights 0.38 for Board and -0.01 for OAS. In 1989 the multiple R is 0.28 and the beta weights 0.24 for Board 0.14 for OAS. None of these measures is statistically significant because of the low number of schools (as opposed to the high number of students). However, the evidence in both 1987 and 1989 points to higher student satisfaction with curriculum choice being more strongly associated with a greater range of Board determined courses, which count towards matriculation, rather than with greater participation in school-based OAS courses. It would certainly be worthwhile extending this research to include a larger sample of schools. The number of Board determined courses on offer does correlate strongly with the size of the school (for 1987 r = .74 p < .01 one-tailed and for 1989 r = .57 p < .01 one-tailed) and therefore with the number of teachers on the school staff. Student Characteristics and Favourable Perceptions of Curriculum Choice Table 4 gives group means for CUR with the groups defined by the student background characteristics already described and used in Table 2. Table 4 Group Means of Students' Satisfaction with Curriculum Choice Group Variable PAR SES GEN NES 1 74.8 (77.4) 1 76.7 (78.8) 1 76.4 (78.4) 1 76.1 (77.8) 2 75.2 (75.7) 2 77.4 (78.4) 2 77.0 (78.4) 2 77.5 (79.6) 3 77.0 (79.5) 3 77.2 (77.4) 3 77.1 (79.6) 4 78.4 (79.9) 4 78.0 (76.9) 4 78.5 (78.4) 5 77.2 (80.0) 6 76.1 (78.8) ETA2 % 0.9 (0.8) 0.1 (0.3) 0.0 (0.0) 0.0 (0.2) p < .01 ns ns ns __________________________________________________________________________ Main figures refer to the 1987 data set; figures in parentheses refer to the 1989 data set. Table 4 shows that none of the student background variables has any substantial influence on students' satisfaction with their curriculum choice. This is to be expected given the regression analyses in Table 2. PAR, parental influence towards higher education, does have a slight influence, and indeed PAR group differences are statistically significant at the p < .01 level. As might be expected, satisfaction with the curriculum choice increases as parental influence towards higher education increases; this is more clearly seen in the 1987 data, where the variation is monotonic. Students were asked about their plans for education after leaving school. Table 5 shows student satisfaction with curriculum choice according to whether the student plans to undertake further study after leaving school. Table 5 Satisfaction with Curriculum Choice According to Further Study Plans Post-school Mean SD N further study plans Satisfaction FS at Univ or equivalent 78.0 (79.1) 17.7 (19.0) 881 (1125) FS elsewhere 75.9 (78.2) 17.1 (18.0) 362 (488) No FS 73.8 (77.0) 18.9 (19.8) 317 (424) ETA2 % 0.9 (0.2) p < .01 (ns) ______________________________________________________________________ Main figures refer to the 1987 data set; figures in parentheses refer to the 1989 data set. Differences among the three groups in Table 5 show a monotonic variation; the greater the propensity towards undertaking post-school education at higher levels, the more likely is a student to be satisfied with the range of subjects offered in the senior school curriculum. This relationship is to be expected for a curriculum primarily suited to matriculation. However, the relationship is not a strong one. ETA2 is small for both the 1987 and 1989 data sets; indeed the variance among the three groups is not statistically significant for the 1989 data. Information was collected from most of the students in 1987 and all of the students in 1989 about their occupational aspirations. Their preferred career choice was classified according to the RIASEC scheme devised by Holland (1985). R (realistic) jobs requiring manual skills I (investigative) jobs where intellectual skills are paramount A (artistic) jobs requiring artistic ability S (social) jobs which directly help people E (entrepreneurial) jobs with managerial, leadership, entrepreneurial roles C (conventional) routine jobs especially clerical While all jobs fall into more than one category, the classification scheme assigns a predominant category, which is used in the following analysis. Table 6 shows mean values of students' perceptions of curriculum choice grouped according to the students' RIASEC classifications. Table 6 Students' Perceptions of Curriculum Choice According to their Career Aspirations Vocational Mean SD N Aspiration CUR Realistic 75.3 (78.6) 18.3 (18.5) 213 (480) Investigative 80.4 (79.8) 16.9 (18.8) 154 (441) Artistic 74.5 (75.8) 17.3 (19.4) 90 (236) Social 78.0 (78.0) 16.7 (18.6) 208 (410) Entrepreneurial 80.3 (79.6) 16.3 (20.2) 66 (205) Conventional 77.3 (79.0) 16.6 (18.9) 79 (209) ETA2 % 1.5 (0.4) p < .05 (ns) __________________________________________________________ Main figures refer to the 1987 data set; figures in parentheses refer to the 1989 data set. Differences among the groups in Table 6 are not large compared to the variance among students. Indeed the variance among groups in the 1989 data set is not statistically significant. However, it is worth noting that the weak patterns which are there show that the curriculum on offer appeals most to students with investigative and entrepreneurial aspirations and least to students with artistic aspirations. This is the sort of pattern one might expect from a curriculum traditionally geared to matriculation requirements. The 1989 data on students' occupational aspirations contain additional coding information. Each student's occupational choice is placed on a sixteen category prestige scale commonly called the ANU scale (Broom et al, 1977) and described earlier in this paper in connection with parental occupations. Unlike the RIASEC scale, the ANU scale describes an ordinal variable, high values on the ANU scale denoting occupations with high connotations of prestige. Patterns of variance of satisfaction with curriculum choice among students grouped by the ANU scale value of their occupational aspirations are discernible only after considerable collapsing of the ANU categories. These patterns are shown in Table 7. Table 7 Students' Perceptions of Curriculum Choice with the Prestige Value of their Career Aspirations Prestige of Career Aspiration CUR SD N 1 (Unskilled to skilled manual) 75.0 19.3 311 2 (Clerical to general administrative) 78.6 19.4 450 3 (Professional) 79.5 18.6 1208 ETA2% 0.7 (p < .01) ___________________________________________________________________ 1989 data. The pattern of variance is not strong but the tendency is for students with higher career aspirations to be more favourably disposed towards the curriculum choice available to them. The Year 11 students surveyed in 1989 belong to the main cohort used in the research, and consequently there are more data on these students. Among the data available are various achievement measures. These additional data allow us to investigate the variance of satisfaction with curriculum choice among students grouped by achievement level. To investigate this variance use was made of the results of statewide reference tests in mathematics and English given to the students in 1988 while they were in Year 10. (The official purpose of these tests is to moderate the school-based student assessments which determine the award of the School Certificate at the end of Year 10). A straightforward addition of the English and mathematics scores was used as a measure of achievement. Scores were collapsed into three categories so that there were roughly equal numbers of students in each category. This high degree of collapsing was used because it allows the weak patterns under investigation to be more clearly illustrated (see Table 8). The 1989 data contained enough information to allow a more refined measure of perceptions of curriculum choice for this analysis. The variable CURmod used in Table 8 is the percentage of positive responses to the subjects studied, where these subjects are weighted according to units allocated to the subject. Units are determined by the timetable periods given to the subject and by the level of difficulty to which the subject is studied. Table 8 Students' Perceptions of Curriculum Choice with Achievement Level Achievement Level CURmod SD N Low 76.1 22.0 550 Intermediate 76.9 20.0 561 High 79.9 18.6 547 ETA2% 0.5 p < .05 ____________________________________________________ 1989 data. CURmod and the achievement score are both interval measures; a straightforward correlation analysis without collapsing the achievement categories yields much the same summary result as the above table; r = +.07, r2 = 0.5%, p < .01 (1-tailed). Table 8 shows a weak pattern in which the higher achieving students are more favourably disposed towards the curriculum choice available to them. Again this is the sort of pattern expected of a curriculum with a traditional orientation towards matriculation requirements. Conclusions Whether or not Year 11 students' have favourable attitudes to their school experience is more strongly influenced by how the students regard their curriculum choices than by student background characteristics. There is some variance among schools in mean levels of satisfaction with curriculum choice, but this variance is much less than the variance among individual students. Schools having higher levels of satisfaction tend to be those which offer a larger number of courses, but it should be noted that this paper considers only the number of courses and does not consider the way in which subject choice is presented to students. Student background characteristics offer little explanation of why some students are more satisfied with curriculum choice than others, except perhaps the extent to which their parents encourage them towards post-school education. Students' own career aspirations offer a slightly better explanation of satisfaction with curriculum choice. Students aspiring towards higher education tend to be more satisfied with the subjects available to them; students aspiring towards higher prestige, investigative or entrepreneurial occupations tend to be more satisfied with the subjects available to them than students with aspirations towards lower prestige occupations. The higher achieving students tend to be more satisfied with the curriculum choice on offer to them, but this association is perhaps not as strong as one might expect. References Broom, L., Duncan-Jones, P. Lancaster-Jones, F. and McDonell, P. (1977). Investigating Social Mobility. Canberra: ANU Press. Ainley, J., Foreman, J and Sheret, M. (In the press). Differences Among High Schools in Their Influence on Students to Stay Beyond the Minimum Leaving Age. Journal of Educational Research. Holland, J.L. (1985). The Self-Directed Search. Odessa, Florida, USA: PAR Inc. Williams, T. and Batten, M. (1981). The Quality of School Life (ACER Research Monograph No.12). Hawthorn, Victoria, Australia: ACER. APPENDIX A SCHOOL LIFE This question contains 40 items on this page and the next page. Each item says that My School is a Place Where some particular thing happens to you or you feel a particular way. We want you to say whether you Strongly Agree, Agree, Disagree, or Strongly Disagree with each item. Please read each item carefully and for each item tick one answer which best describes how you feel. Don't forget that you have to put "My School is a Place Where" before each item for it to make sense. For example, "My School is a Place Where I feel important". MY SCHOOL IS A PLACE WHERE ... (Tick one box in each line) Strongly Strongly Agree Agree Disagree Disagree 1 .. teachers treat me fairly in class     2 .. I feel proud to be a student     3 .. the things I learn are important to me     4 .. people look up to me     5 .. I feel depressed     6 .. I find it easy to get to know other people     7 .. I achieve a standard in my work which I consider satisfactory     8 .. I like learning     9 .. I get enjoyment from being there     10 .. other students are very friendly     11 .. I feel restless     12 .. teachers give me the marks I deserve     13 .. I have acquired skills that will be of use to me when I leave school     14 .. I always achieve a satisfactory standard in my work     15 .. other people care what I think     16 .. teachers take a personal interest in helping me with my school work     17 .. people trust me     c ACER 1987 MY SCHOOL IS A PLACE WHERE ... (Tick one box in each line) Strongly Strongly Agree Agree Disagree Disagree 18 .. mixing with other people helps me to understand myself     19 .. I feel lonely     20 .. the things that I learn will help in my adult life     21 .. I know people think a lot of me     22 .. I know how to cope with the work     23 .. teachers help me to do my best     24 .. I get upset     25 .. I am given the chance to do work that really interests me     26 .. I know I can do well enough to be successful     27 .. the things I am taught are worthwhile learning     28 .. I feel important     29 .. teachers are fair and just     30 .. I am a success as a student     31 .. I really like to go each day     32 .. I learn to get along with other people     33 .. I feel worried     34 .. the work I do is good preparation for my future     35 .. I feel proud of myself     36 .. other students accept me as I am     37 .. I have learnt to work hard     38 .. I get on well with the other students in my class     39 .. I find that learning is a lot of fun     40 .. teachers listen to what I say     c ACER 1987 APPENDIX B CURRICULUM CHOICE Please list the subjects you have studied this year. You may list as many as ten subjects or, perhaps less than ten. For each subject tick either box A or box B, whichever box you think more accurately describes your feelings about the subject. Box A means: I am doing this subject because it is something I want to do, or it is my choice, or I think it will be useful for me. Box B means: I am only doing this subject because it is compulsory, or I was told I ought to do it, or it was needed before I could do other subjects, or it was the one available on the timetable. There is also a space against each subject if you would like to make a comment on that subject. Subject Box A Box B Comment (optional) 1 ___________________________   _________________________________ 2 ___________________________   _________________________________ 3 ___________________________   _________________________________ 4 ___________________________   _________________________________ 5 ___________________________   _________________________________ 6 ___________________________   _________________________________ 7 ___________________________   _________________________________ 8 ___________________________   _________________________________ 9 ___________________________   _________________________________ 10 ___________________________   _________________________________ c ACER 1987