High Educ DOI 10.1007/s10734-009-9200-6
Undergraduates’ learning experience and learning process: quantitative evidence from the East Beverley J. Webster Æ Wincy S. C. Chan Æ Michael T. Prosser Æ David A. Watkins
Ó Springer Science+Business Media B.V. 2009
Abstract This article examines the construct validity of the Course Experience Questionnaire (CEQ) in Hong Kong and investigates the similarities and differences in the process of learning among students in different disciplinary studies. It is based on a survey of 1,563 undergraduate students in two disciplines, humanities and sciences, and of principally Chinese ethnicity. Findings from exploratory and confirmatory factor analyses support the scale structure of the four subscales of a modified version of the CEQ (good teaching, clear goals and standards, appropriate work, and appropriate assessment) in a non-Western context and could provide a basis for cross cultural research and international benchmarking. While there was variation across subgroups, there was a genuine pattern of relationships between the perceptions of learning environment and learning strategies shown by structural modeling. This information could be used to inform the design of discipline-specific programs in the new curriculum. Keywords Course experience Learning strategy Undergraduates Hong Kong Chinese
Introduction There is an increasing number of student surveys of perceptions of university teaching and learning environments and experiences. Surveys of this kind are now commonplace in countries such as Australia and the UK. They are being utilized for reasons of either accountability or learning improvement or sometimes both. In such contexts, the Course Experience Questionnaire (CEQ; Ramsden 1991; Wilson et al. 1997) has been widely used. For example, all graduates in Australia are asked to complete the CEQ as part of a graduate destination questionnaire. The Australian Government has used and is still using B. J. Webster W. S. C. Chan (&) M. T. Prosser Centre for the Advancement of University Teaching, University of Hong Kong, Pokfulam, Hong Kong e-mail:
[email protected] D. A. Watkins Faculty of Education, University of Hong Kong, Pokfulam, Hong Kong
123
High Educ
such data to assess the performance and ascertain the needs of its universities. Most Australian universities have also analyzed and reported their own institution’s CEQ scores for internal purposes, often conducting surveys of their own undergraduates to inform curriculum and staff development efforts. The CEQ evolved from theory and continued research on student learning in higher education over the past few decades and it is believed that the information obtained from the survey can be useful for professional development. Survey results can highlight issues to address in relation to the student learning experience. Such issues we know are related to the promotion of deeper learning approaches in students which are necessary for better learning outcomes which are desired at tertiary level. Universities in Hong Kong are now being asked to demonstrate the quality of the learning outcomes of their students for system wide accountability and improvement purposes (University Grants Committee 2005). The CEQ would seem to be a potentially useful instrument upon which to base evidence to such policy and practice. However, as yet there is little evidence at the institution level of the reliability and validity of the CEQ in a non-Western context such as Hong Kong. There is also little evidence from anywhere justifying the construct validity and the stability of the CEQ factors for students of different academic years or disciplines assumed in such surveys (see Ginns et al. 2007). The purpose of this paper is to provide such evidence for a Hong Kong university.
The Course Experience Questionnaire The CEQ was designed from within the well-known student learning perspective. The present form of the CEQ originated in the qualitative work of Marton and Sa¨ljo¨ (1976) and elaborated by others quantitatively such as Entwistle and Ramsden (1983). From this perspective, university students’ approaches to study are contingent upon both their prior experiences of teaching and learning and their perceptions of their current teaching and learning environment. Students have been shown to adopt either a surface approach to study, focusing on short term reproduction, or a deep approach focusing on longer term understanding. Their perceptions of the quality of teaching, the clarity of goals and standards, whether the workload is so high they cannot understand it all, and whether their assessments test reproductive learning rather than understanding have been shown to relate to these approaches to learning (Biggs and Tang 2007; Prosser and Trigwell 1999). There is a substantial body of literature confirming the factor structure of the CEQ within teaching and learning contexts in the West (Byrne and Flood 2003; Lizzio et al. 2002; Ramsden 1991; Richardson 1994, 2005a, b; Sadlo 1997; Trigwell and Prosser 1991b; Wilson et al. 1997). The construct validity of the CEQ was demonstrated by Richardson (2006) by the fact that the scales used in that study collectively defined a single higher-order factor that could be interpreted as a measure of perceived academic quality. Within those contexts the CEQ has been used for a range of different purposes including benchmarking, as a performance indicator, and for summative funding related and formative purposes. For example, it is used by Australian universities to benchmark their students’ learning experiences against counterparts locally, and more recently, internationally (Ginns et al. 2007). More recently, the Australian Government has been using the results as one part of a basket of performance indicators in its performance based funding model, allocating substantial funding to universities performing well in terms of these indicators. Finally, it is used by individual universities to formatively
123
High Educ
evaluate and improve their undergraduate programs (Barrie et al. 2005) and because there is evidence showing that students are best placed to evaluate many aspects of teaching, and their own ratings are valid, multidimensional and reliable (Marsh 1987; Watchel 1998), course experience can be considered to be quite strongly related to qualities of the actual study context.
A non-Western context Leading researchers have long warned about the dangers of assuming that theories developed and research conducted about affective and cognitive processes in one culture are appropriate for another (Boekaerts 2003; Markus and Kitayama 1991). However, the general principles of the research and theorizing about student approaches to learning as measured by the Study Process Questionnaire (SPQ; Biggs 1987) and the Approaches to Studying Inventory (Entwistle and Ramsden 1983) have been shown to be valid for Hong Kong Chinese students (Biggs 1992; Kember and Gow 1990). In a cross-cultural meta-analysis which included data from four samples of Hong Kong secondary and tertiary students showed that, similar to Western studies reported above, surface learning approaches were consistently related to learning environments where the students perceived the workload and assessment to be inappropriate. On the other hand, deep level approaches were associated with environments where the teaching was seen as good and the teachers supportive (Watkins 2001). A study with an earlier version of the CEQ did provide evidence of its reliability for Hong Kong university students (Ho 1998). More recently, in a cross-cultural study and using a revised SPQ, factorial invariance was determined between samples of students in universities in Australia and Hong Kong (Leung et al. 2008). The researchers concluded that more research was needed to determine whether the identified configural invariance results were applicable across fields of study in addition to examining the relations between approaches to learning and perceptions of the learning environment using structural models. The CEQ has recently been used at a university in Hong Kong in an investigation of the validity of adopting the CEQ as one of the key performance indicators and using the data to benchmark with other universities internationally. The intention was to use the evidence obtained from the survey to support and monitor the 4-year undergraduate curriculum which would be implemented in the next few years.
Aims of study The aims of this study were to provide evidence of the CEQ when used with Hong Kong Chinese undergraduates of: (1) (2) (3)
The goodness of fit and reliability of the data to the hypothesized scale structures; The construct validity in terms of relationships between perceptions of course experience and learning strategies; and The baseline structure (Byrne et al. 1989) of these relationships by discipline area and year of study.
123
High Educ
Methodology Sample The original sample consisted of 1988 undergraduate students enrolled in either year 1 or year 3 of study in the academic year 2006–2007 at a university in Hong Kong. The sample came from all 10 faculties and represented approximately 33% of the population. This data included students who were Chinese and who spoke Chinese as their first language. Respondents enrolled in the Faculty of Architecture were subsequently excluded in view of the speculated anomalies in the data from this Faculty (further details are provided as follows). The final sample thus consisted of 1,536 Chinese undergraduate students with a mean age of 20.7 years (SD = 1.58) and of whom 848 were female and 688 were male. The sample was subsequently divided into two broad disciplines of study: the Humanities (n = 688; including students enrolled in the Faculties of Arts, Business and Economics, Education, Law, and Social Sciences) and the Sciences (n = 848; including respondents from the Faculties of Dentistry, Engineering, Medicine, and Science). Please note that the similarities in these numbers were purely coincidental. Data collection and instruments The survey was available in both online and paper versions. Invitations were sent via the University email accounts and visits to lectures, libraries, and examination halls at the end of the academic year. Participation was voluntary. Ethics approval was obtained prior to commencement of the study. Seventeen CEQ items, corresponding to Good Teaching, Clear Goals and Standards, Appropriate Assessment, and Appropriate Workload scales, were adapted from the University of Sydney Student Course Experience Questionnaire (SCEQ; University of Sydney, Institute for Teaching and Learning 2005). The Sydney SCEQ was more suitable for this study as it was shortened, modified, and validated for current undergraduate students (Ginns et al. 2007). The students responded to each of these 17 items by indicating their agreement or disagreement with a particular statement along a 5-point scale (1 = strongly disagree to 5 = strongly agree). Prior to the mainframe data collection, the SCEQ items were piloted among undergraduate students who came from different disciplines. These items were slightly modified to be applicable to the Hong Kong context; for example, a ‘degree curriculum’ was used instead of a course or degree course. Items of the current SCEQ version are listed in Table 1. Reponses to items with negative wordings (marked ‘**’ in Table 1) were coded in reverse before calculating the scale scores. Fourteen learning strategy items from the SPQ (Biggs 1987) were also used in the study to assess students’ learning strategies as either deep or surface. Students responded to these 14 items on a 5-point scale to statements related to how they went about their study (1 = this is never true for me to 5 = this is always true for me). Additional data that provided some background information on the participants was also collected. Analysis Exploratory factor analysis using SPSS 15.0 (Chicago, IL) was initially conducted to test the structure of the 17 SCEQ items. This included analysis involving 10 Faculties by discipline area (humanities and sciences) and year of study (year 1 and year 3). As a result
123
High Educ Table 1 Factor loadings of 17 SCEQ items by principal component analysis Scales
Items
Good teaching
1. The teachers normally give me helpful feedback on my progress
0.669
0.174
0.146
-0.073
2. The teachers of the degree curriculum motivate me to do my best work
0.694
0.211
0.140
-0.112
3. The staff make a real effort to understand difficulties I may be having with my work
0.683
0.103
4. My lecturers are extremely good at explaining things
0.727
0.093
5. The teachers work hard to make their subjects interesting
0.737
6. The staff put a lot of time into commenting on my work
0.660
-0.054
-0.125
1. I have usually had a clear idea of where I am going and what is expected of me in this degree curriculum
0.305
0.592
-0.054
2. It is always easy to know the standard of work expected
0.315
0.593
-0.228
3. The staff made it clear right from the start what they expected from students
0.509
0.345
-0.121
4. It has often been hard to discover what is expected of me in this degree curriculum**
-0.132
0.747
0.312
Clear goals and standards
Appropriate assessment
2
3
4
0.128
0.077 0.052
1. The staff seem more interested in testing what I have memorised than what I have understood**
0.766
2. Too many staff ask me questions just about facts**
0.664
3. To do well in this degree all you really need is a good memory** Appropriate workload
1
0.100
0.667
1. There is a lot of pressure on me as a student in this degree curriculum**
0.103
0.142
0.170 0.795
2. The workload is too heavy**
0.083
3. I am generally given enough time to understand the things I have to learn
0.410
0.167
-0.164
0.460
4. The volume of work necessary to complete this degree curriculum means it cannot all be thoroughly comprehended**
-0.149
0.099
0.185
0.451
0.807
Eigenvalues
3.58
1.80
1.78
1.52
% Variance (50.98)
21.07
10.58
10.42
8.91
** Reversed items Notes: (i) Figures in bold indicate factor loadings on a priori factors. (ii) Figures in italics indicate a cross loading of 0.3 or higher
123
High Educ
of this initial analysis, cases from the Faculty of Architecture were dropped due to the anomalies in data potentially explained by the two distinct programs, one humanitiesrelated and the other science-related, offered by the Faculty. As the survey did not ask the students to indicate the program they were enrolled in, it was not possible to separate these students into disciplinary groups and thus they were excluded from subsequent analyses. Confirmatory factor analysis was conducted using LISREL 8.8 (Jo¨reskog and So¨rbom 1996). The whole sample was randomly split into two groups. The first sample (n = 751) was used for the confirmatory analysis process where re-specification and estimation of the data were conducted to result in the best overall model fit of the scale structures of both the SCEQ and the SPQ. The second sample (n = 785) was used for the validation of the model as recommended by Anderson and Gerbing (1988). The use of confirmatory factor analysis for final reporting of results was appropriate in this study as a priori measurement models were tested and factor structures in different homogenous subgroups of the sample was examined (Watkins 1989). A weighted least square (WLS) approach was used to estimate the goodness-of-fit indices between the data and the specified models based on an asymptotic covariance matrix. WLS is used to provide asymptotic unbiased parameter estimates with ordinal observed variables on large sample sizes (Boomsma and Hoogland 2001; Jo¨reskog 1994). Multivariate non-normality of the variables was checked by treating the variables as continuous and by looking at the distributions of the ordinal data. According to Jo¨reskog and So¨rbom (1996), the minimum sample size for WLS is (k ? 1)(k ? 2)/2 where k is the number of indicators in the model (Flora and Curran 2004). The largest number of indicators we used in a single model was 17 (from 17-item SCEQ), which would mean the minimum sample size required was 171, which is smaller than the size of any subgroup being tested in the study. Assessment of model fit included the conventional v2 statistics, which is preferably not significant; the root mean square error of approximation (RMSEA; Steiger 1990), for which values below 0.05 indicate good fit and values as high as 0.08 indicate moderate fit; and the comparative fit index (CFI; Bentler 1990), non-normed fit index (NNFI; Bollen 1989), and the adjusted goodness of fit index (AGFI; Jo¨reskog and So¨rbom 1982), for which values [0.90 indicate good fit. RMSEA, expressed in per degree of freedom, compares the fit by estimating the discrepancy between the testing and hypothesized models based on the non-centrality parameter. The AGFI takes into account both the sample size and number of parameters in the estimation of model fit. The NNFI compares the fit of the two models, and the CFI compares the non-central v2 to the null model. After the confirmatory and validation analysis on six-one-factor congeneric models (four SCEQ and two SPQ models), a composite score was calculated for each of the six scales based on factor score regression weights produced in the LISREL output estimates using a non-unit weighted score which reflected the actual contribution each item made to the scales (Rowe et al. 1995). A measure of composite reliability (rc) was estimated for each of the six scales using WLS regression estimates and error variance estimates from the LISREL output (Rowe 2006; Tarkkonen and Vehkalahti 2005). Taking into account the unidimensionality of the SCEQ and SPQ scales, constructing composites based on a priori questionnaire construction was a proper approach to minimize unwanted sources of variance in arriving at the model solution (Little et al. 2002). Structural equation modeling was then used to test the baseline structure of relationships of perceptions of course experience and learning strategies and the potential differences by both discipline area and year of study. These baseline structures tested on the four subgroups were then compared.
123
High Educ
Results Exploratory factor analysis A principal component analysis using a Varimax rotation method of the 17-item SCEQ produced a four factor solution based on the eigenvalue [ 1 criterion, which accounted for 51.0% of the variance. When the factor loading cut off was set at 0.3, items related to clear goals and standard cross loaded with items related to good teaching and one item from the appropriate workload scale also loaded on the good teaching scale (see Table 1). With minor variations, this structure was similar for each Faculty by year of study. For some of these analyses, the small sample size could explain minor deviations from the identified structure. The scale structure of the 14 SPQ items was also explored and also indicated similar results as previously identified in Chinese undergraduate students. The results show a clear two-scale structure in terms of factor loading; however, the percentage of variance explained by these two factors was low (35.3%). The surface strategy items loaded together on one factor with factor loadings [ 0.4. Deep strategy items loaded on another factor while one of these items (While I am studying, I think of real life situations to which the material that I am learning would be useful.) cross loaded with surface strategy items. The cumulative variance was 20.7% for the first factor (eigenvalue = 2.89) and 14.6% for the second factor (eigenvalue = 2.05). As with the SCEQ scale structures, the SPQ scale structures were similar by faculty and year of study with minor variations. Goodness-of-fit of measurement models Good fit estimates were identified for the four one-factor congeneric SCEQ measurement models, good teaching, clear goals and standards, appropriate workload, and appropriate assessment, and the two one-factor congeneric SPQ models, deep strategy and surface strategy (see Table 2). The significant covariances between pairs of independent variables in the models were being specified (Byrne 1998) and the number of parameters being observed was explained by the degrees of freedom. The Chi-squares for each scale in both the SCEQ and the SPQ were small and not significant (P [ 0.05); the RMSEA values were \0.05; and the NNFI, the DFI, and the AGFI were all estimates close to 1.00 indicating a good fit to the model for each of these six scales. The composite reliabilities (rc) estimates indicated a good reliability for most scales. The exceptions were surface learning strategies (rc = 0.541) and clear goals and standards (rc = 0.575).
Table 2 Fitted one-factor congeneric models for SCEQ and SPQ: goodness of fit summary and composite reliabilities Composite variable
v2
df
P
RMSEA
NNFI
CFI
AGFI
rc
Deep strategy
15.78
12
0.20
0.020
0.99
0.99
0.99
0.757
Surface strategy
16.46
10
0.09
0.029
0.95
0.97
0.99
0.541
Good teaching
13.62
7
0.06
0.036
0.98
0.99
0.99
0.837
Clear goals & standards
5.43
2
0.07
0.048
0.91
0.97
0.99
0.575
Appropriate assessment
1.04
1
0.31
0.007
1.00
1.00
1.00
0.794
Appropriate workload
1.68
1
0.20
0.001
0.99
1.00
0.99
0.620
123
High Educ
The 17 SCEQ items and 14 SPQ items were subsequently tested in two measurement models. The four-factor SCEQ model (v2 = 605.95; df = 113; P [ 0.05; RMSEA = 0.076; NNFI = 0.75; CFI = 0.80; AGFI = 0.95) and the two-factor SPQ model (v2 = 477.54; df = 76; P [ 0.05; RMSEA = 0.084; NNFI = 0.61; CFI = 0.68; AGFI = 0.94) did not, unsurprisingly, fit the data as well as the single congeneric models. Structural model of relationships of perceptions of course experiences on deep and surface learning strategies Structural models were produced to examine the overall model fit and relative contribution of each of the four SCEQ scales to learning strategies on four subgroups of the sample: Humanities year 1 (H1), Humanities year 3 (H3), Sciences year 1 (S1), and Sciences year 3 (S3). The goodness-of-fit indices indicated that all models were a good fit to the data (see Fig. 1). Chi-square estimates were small and not significant; the RMSEA values were \0.05; and the NNFI, the DFI, and the AGFI were all estimated close to or equal to 1.00. The results showed that for all four subgroups, with the exception of year 3 Humanities, perception of good teaching and perception of clear goals and standards were associated with deep learning strategies and all of these paths were significant at the 95% level. For all models except year 3 Humanities, perception of the appropriateness of assessment affected
Fig. 1 Structural models of the effects of course experiences on learning strategies for four groups of students
123
High Educ
surface learning strategies negatively. In all four models the perception that the workload was inappropriate was associated with surface learning while for year 1 Sciences it also affected deep learning. For all subgroups but year 1 the perception of good teaching was also related to surface learning strategies.
Discussion Indicators of the quality of teaching and learning in higher education are constantly sought after as governments, employers, and the public concentrate on measuring accountability and demand quality outcomes. One such indicator of a quality outcome is that students are adopting deeper learning strategies since it is well known that this leads to a better understanding of the curriculum and a better overall learning experience (Biggs 1993). Knowing the contribution that student perceptions of their learning environment can make to learning strategies is seen as important in improving learning outcomes. This discussion is structured around three highlights from the results of this study. First, there is the confirmation of the scale structures of 17 SCEQ items and 14 SPQ items and the identified anomalies. Second is the construct validity of the relationships between perceptions of course experience and learning strategies. Lastly, it is the stability of baseline structures across discipline area and year of study. The results provide support for the scale structure with this Chinese undergraduate sample regardless of discipline area (Humanities and Sciences) or year of study (year 1 and year 3). Although the initial factor analysis showed that the 17 course experience items formed a four-scale structure, there were several items from the good teaching scale that cross loaded on the clear goals and standards scale. In previous studies, items from the good teaching scale were loaded on two scales (Kreber 2003) or loaded with appropriate assessment items (Wilson et al. 1997). In this study although there were cross loadings, the highest loadings were on the good teaching scale. Subsequent to this analysis, the estimates from the confirmatory factor analysis showed good fit to the four-one-factor congeneric measurement models for the SCEQ. All Chi-square estimates were small and not significant; the RMSEA values \0.05; and the NNFI, the DFI, and the AGFI estimates were all close to 1.00. The composite reliabilities were between 0.575 and 0.837. These results indicate that the scale structure of the 17 SCEQ items was working to some extent in undergraduate Hong Kong Chinese students. The SPQ has previously been validated in this population (Biggs 1992; Kember and Leung 1998; Watkins 2001) and fit estimates reported in this paper were similar to these previous studies. It is noted that the two-factor structure was not a simple structure nor was the reliability estimate for the surface strategy scale very high (rc = 0.541). The four-factor congeneric measurement model of the SCEQ did not fit the data as well. However, it is said to be unrealistic to find well fitting hypothesized models such as these where v2/df is not significant (Byrne 2001) and these estimates are similar to those found by Diseth et al. (2006). Evidence of the construct validity of the SCEQ was obtained by achieving good fit estimates from an examination of the relationships between the SCEQ scales and the SPQ scales on structural models by discipline area (Humanities and Sciences) and year of study (year 1 and year 3). An investigation of the contribution of student perceptions of learning environment to learning strategies in subgroups of students established evidence of stable and well-fit hypothesized models for all year 1 students and year 3 Science students. That is, students who perceived the teaching as being good and the goals and standards to be clear in the
123
High Educ
degree curriculum were also those students who adopted deeper learning strategies. Among such students, those who perceived the workload and assessment as appropriate were less likely to adopt surface learning strategies. Although the path estimates for the year 3 Humanities group for all of these were in the same direction, they were small and not significant with one exception: year 3 Humanities students who perceived workload as appropriate were also less likely to adopt surface learning strategies. An interesting finding in these data was that for all Science students and year 3 Humanities students, those who perceived the teaching as good were also more likely to adopt surface learning strategies. This relationship could be interpreted as Chinese teachers giving only factual information which would lead to rote learning. This speculation supports the reputed tendency of Chinese teachers toward spoon-feeding their students or that of Chinese students to preferring to be spoon-fed (Kember 2000). Nonetheless, the relationship between good teaching and both deep and surface learning strategies could be better explained by the nature of understanding from the perspective of Chinese learners. Previous in-depth qualitative study on the relationship of memorization and understanding suggests that for Chinese learners memorizing the information as the first step could enhance subsequent deep understanding of the content (Kember and Gow 1991). While the same relationship was not evident in year 1 Humanities students, it is speculated that the first year curriculum of Humanities subjects emphasize a broad spectrum of general knowledge which demands less content-specific knowledge such as terminologies and professional skills which to certain extent requires memorization. Another interesting finding that emerges from this study is that for year 3 Science students the perception of inappropriate workload was associated with both surface and deep learning strategies. For Hong Kong students, science assignments since senior high school levels emphasize critical thinking and the application of theories to real life practice. There is a perception that so much work that students cannot possibly get through could induce rote learning (Trigwell and Prosser 1991a), as it seems the only way to cope with the perceived overload is to memorize. However, the actual assignment tasks might also stimulate deep understanding of the content especially for senior year Science students. With the evidence on construct validity and stable baseline structures among different subgroups of Hong Kong Chinese undergraduate students, the SCEQ could be a reliable instrument for the evaluation of effectiveness of higher education in Hong Kong in terms of teaching quality, the clarity of goals and standards, and the appropriateness of assessment and workload. While at the same time the relationships between perceptions of course experience and learning strategies varied among subgroups of students, these differences could inform the specific needs of degree courses in the design of the new curriculum. Adopting the SCEQ in Hong Kong universities could also provide a basis for cross cultural research and international benchmarking purposes in the future.
References Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411–423. doi:10.1037/0033-2909.103. 3.411. Barrie, S. C., Ginns, P., & Prosser, M. (2005). Early impact and outcomes of an institutionally aligned, student focused learning perspective on teaching quality assurance. Assessment & Evaluation in Higher Education, 30(6), 641–656. doi:10.1080/02602930500260761. Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107(2), 238– 246. doi:10.1037/0033-2909.107.2.238.
123
High Educ Biggs, J. B. (1987). Student approaches to learning and studying. Hawthorn, VC: Australian Council for Educational Research. Biggs, J. B. (1992). Why and how do Hong Kong students learn? Using the learning and study process questionnaires (education paper No. 14). Hong Kong: The University of Hong Kong, Faculty of Education. Biggs, J. B. (1993). What do inventories of students’ learning processes really measure? A theoretical review and clarification. The British Journal of Educational Psychology, 63(1), 3–19. Biggs, J. B., & Tang, C. (2007). Teaching for quality learning at university (3rd ed.). Maidenhead, Berkshire: McGraw-Hill Education. Boekaerts, M. (2003). How do students from different cultures motivate themselves for academic learning? In F. Salili & R. Hoosain (Eds.), Teaching, learning, and motivation in a multicultural context (pp. 13– 32). Greenwich, CT: Information Age Publishing Inc. Bollen, K. A. (1989). A new incremental fit index for general structural equation models. Sociological Methods & Research, 17(3), 303–316. doi:10.1177/0049124189017003004. Boomsma, A., & Hoogland, J. J. (2001). The robustness of LISREL modeling revisited. In R. Cudeck, S. du Toit, & D. So¨rbom (Eds.), Structural equation modeling: Present and future (pp. 1–25). Lincolnwood, IL: Scientific Software International. Byrne, B. M. (1998). Structural equation modeling with LISREL, PRELIS, and SIMPLIS: Basic concepts, applications, and programming. Mahwah, NJ: Lawrence Erlbaum Associates. Byrne, B. M. (2001). Structural equation modeling with AMOS, EQS, and LISREL: Comparative approaches to testing for the factorial validity of a measuring instrument. International Journal of Testing, 1(1), 55–86. doi:10.1207/S15327574IJT0101_4. Byrne, B. M., & Flood, B. (2003). Assessing the teaching quality of accounting programs: An evaluation of the course experience questionnaire. Assessment & Evaluation in Higher Education, 28(2), 135–145. doi:10.1080/02602930301668. Byrne, B. M., Shavelson, R. J., & Muthe´n, B. (1989). Testing for the equivalence of factor covariance and mean structures: The issue of partial measurement invariance. Psychological Bulletin, 105(3), 456– 466. doi:10.1037/0033-2909.105.3.456. Diseth, A., Pallesen, S., Hovland, A., & Larsen, S. (2006). Course experience, approaches to learning and academic achievement. Education & Training, 48(2/3), 156–169. doi:10.1108/00400910610651782. Entwistle, N. J., & Ramsden, P. (1983). Understanding student learning. London and Canberra: Croom Helm. Flora, D. B., & Curran, P. J. (2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological Methods, 9(4), 466–491. doi: 10.1037/1082-989X.9.4.466. Ginns, P., Prosser, M., & Barrie, S. (2007). Students’ perceptions of teaching quality in higher education: The perspective of currently enrolled students. Studies in Higher Education, 32(5), 603–615. doi: 10.1080/03075070701573773. Ho, A. S. P. (1998). Changing teachers’ conceptions of teaching as an approach to enhancing teaching and learning in tertiary education. Unpublished doctoral thesis, The University of Hong Kong, Hong Kong. Jo¨reskog, K. G. (1994). Structural equation modeling with ordinal variables. In T. W. Anderson, K. T. Fang, & I. Olkin (Eds.), Proceedings of the international symposium on multivariate analysis and its applications. Multivariate analysis and its applications (pp. 297–310). Hayward, CA: Institution of Mathematical Statistics. Jo¨reskog, K. G., & So¨rbom, D. (1982). Recent developments in structural equation modeling. Journal of Marketing Research, 19(4), 404–416. doi:10.2307/3151714. Jo¨reskog, K. G., & So¨rbom, D. (1996). LISREL 8: User’s reference guide. Lincolnwood, IL: Scientific Software International, Inc. Kember, D. (2000). Misconceptions about the learning approaches, motivation and study practices of Asian students. Higher Education, 40(1), 99–121. doi:10.1023/A:1004036826490. Kember, D., & Gow, L. (1990). Cultural specificity of approaches to study. The British Journal of Educational Psychology, 60(3), 356–363. Kember, D., & Gow, L. (1991). A challenge to the anecdotal stereotype of the Asian student. Studies in Higher Education, 16(2), 117–128. doi:10.1080/03075079112331382934. Kember, D., & Leung, D. Y. P. (1998). The dimensionality of approaches to learning: An investigation with confirmatory factor analysis on the structure of the SPQ and LPQ. The British Journal of Educational Psychology, 68(3), 395–407. Kreber, C. (2003). The relationship between students’ course perception and their approaches to studying in undergraduate science courses: A Canadian experience. Higher Education Research & Development, 22(1), 57–75. doi:10.1080/0729436032000058623.
123
High Educ Leung, D., Ginns, P., & Kember, D. (2008). Examining the cultural specificity of approaches to learning in universities in Hong Kong and Sydney. Journal of Cross-Cultural Psychology, 39(3), 251–266. doi: 10.1177/0022022107313905. Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9(2), 151–173. doi:10.1207/ S15328007SEM0902_1. Lizzio, A., Wilson, K., & Simons, R. (2002). University students’ perceptions of the learning environment and academic outcomes: Implications for theory and practice. Studies in Higher Education, 27(1), 27–52. doi:10.1080/03075070120099359. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98(2), 224–253. doi:10.1037/0033-295X.98.2.224. Marsh, H. W. (1987). Students’ evaluations of university teaching: Research findings, methodological issues, and directions for future research. International Journal of Educational Research, 11(3), 253–388. doi:10.1016/0883-0355(87)90001-2. Marton, F., & Sa¨ljo¨, R. (1976). On qualitative differences in learning: I-Outcome and process. The British Journal of Educational Psychology, 46(1), 4–11. Prosser, M., & Trigwell, K. (1999). Understanding learning and teaching: The experience in higher education. Philadelphia, PA: Open University Press. Ramsden, P. (1991). A performance indicator of teaching quality in higher education: The course experience questionnaire. Studies in Higher Education, 16(2), 129–150. doi:10.1080/03075079112331382944. Richardson, J. T. E. (1994). Gender differences in mental rotation. Perceptual and Motor Skills, 78(2), 435–448. Richardson, J. T. E. (2005a). Instruments for obtaining student feedback: A review of the literature. Assessment & Evaluation in Higher Education, 30(4), 387–415. doi:10.1080/02602930500099193. Richardson, J. T. E. (2005b). Students’ approaches to learning and teachers’ approaches to teaching in higher education. Educational Psychology, 25(6), 673. doi:10.1080/01443410500344720. Richardson, J. T. E. (2006). Investigating the relationship between variations in students’ perceptions of their academic environment and variations in study behaviour in distance education. The British Journal of Educational Psychology, 76(4), 867–893. doi:10.1348/000709905X69690. Rowe, K. J. (2006). Practical multilevel analysis with MLwiN & LISREL: An integrated course (5th ed., rev.). 22nd ACSPRI summer program in social research methods and research technology, Australian National University, 23–27 January 2006. Camberwell, VIC: Australian Council for Educational Research. Rowe, K. J., Hill, P. W., & Holmes-Smith, P. (1995). Methodological issues in educational performance and school effectiveness research: A discussion with worked example. Australian Journal of Education, 39(3), 217–248. Sadlo, G. (1997). Problem-based learning enhances the educational experiences of occupational therapy students. Education for Health, 10(1), 101–114. Steiger, J. H. (1990). Structural model evaluation and modification: An interval estimation approach. Multivariate Behavioral Research, 25(2), 173–180. doi:10.1207/s15327906mbr2502_4. Tarkkonen, L., & Vehkalahti, K. (2005). Measurement errors in multivariate measurement scales. Journal of Multivariate Analysis, 96(1), 172–189. doi:10.1016/j.jmva.2004.09.007. Trigwell, K., & Prosser, M. (1991a). Improving the quality of student learning: The influence of learning context and student approaches to learning on learning outcomes. Higher Education, 22(3), 251–266. doi:10.1007/BF00132290. Trigwell, K., & Prosser, M. (1991b). Relating approaches to study and quality of learning outcomes at the course level. The British Journal of Educational Psychology, 61(3), 265–275. University Grants Committee. (2005). Education quality work: The Hong Kong experience. Hong Kong: The Hong Kong Polytechnic University, Educational Development Centre. University of Sydney, Institute for Teaching and Learning.(2005). 2005 Student course experience questionnaire (SCEQ). Retrieved December 4, 2006, from http://www.itl.usyd.edu.au/sceq2005. Watchel, H. (1998). Student evaluation of college teaching effectiveness: A brief review. Assessment & Evaluation in Higher Education, 23(2), 191–211. doi:10.1080/0260293980230207. Watkins, D. (1989). The role of confirmatory factor analysis in cross-cultural research. International Journal of Psychology, 24(1), 685–701. doi:10.1080/00207598908246806. Watkins, D. (2001). Correlates of approaches to learning: A cross-cultural meta-analysis. In R. Sternberg & L. Zhang (Eds.), Perspectives on thinking, learning, and cognitive styles (pp. 165–195). Mahwah, NJ: Lawrence Erlbaum Associates. Wilson, K. L., Lizzio, A., & Ramsden, P. (1997). The development, validation and application of the course experience questionnaire. Studies in Higher Education, 22(1), 33–53. doi:10.1080/03075079712331381121.
123
