Methods and Techniques
Sample
For secondary analysis, the sample is outlined in the Technical Report of the IEA civic Education Study (Shultz and Sibberns, 2004, p.44). Schools were selected within a stratified sampling frame in the first stage, and single intact classes were selected in the second stage. Two explicit strata were used in the sampling frame: district and financial modes. The number of schools selected within each stratum were proportional to the total number of schools. The target populations within schools will follow the original IEA population targets: the young population were made up of students from 14.00–14.11 (Form 3), and the older population were 17.00–17.11 (Form 6)

Instrumentation
The ACNIS to be developed for the Phase 2 of the study will combine parts of two existing instruments: the IEA survey (Schultz and Sibberns, pp.242–281) and the National Attitudes Questionnaire (Fairbrother, 2003), both of which have already been translated into Chinese. The ACNIS will be piloted with a small sample of schools and finalised based on the Item Response Theory item analysis to test the function of items to discriminate amongst individuals.

Data collection
ACNIS will be distributed to schools at a time most suitable for its completion by all students within a common time frame. Completion time is estimated at one full class period. Project staff will pick up the completed surveys.

Data Analysis
Data Analysis Framework

References:

1. Schulz, W., & Sibberns, H. (Eds.). (2004). Technical Report on the IEA Civic Education Study. Amsterdam: International Association for Educational Assessment.
2. 2. Fairbrother, G. P. (2003). Toward Critical Patriotism: Student Resistance to Political Education in Hong Kong and China. Hong Kong: Hong Kong University Press.
   

• Latent structures
• Causal modelling
• Multiple methodologies

1. Latent structures
   We have shown that the latent structures of such key constructs as “attitudes to the good citizen” and ‘”political trust” differ from those identified in the original CivEd study (Kennedy, 2010, Kennedy, Huang & Chow, 2011). The key difference between the original study and this project is the sample. CivEd used composite international samples (n = 500 from the 28 participating countries), whereas the current study used Hong Kong samples only. The results provide important justification for secondary analysis, for using local rather than composite international samples, and for developing a more authentic picture of Hong Kong students’ attitudes toward citizenship. This does not imply that the results of the original study were wrong. Data are capable of exhibiting multiple structures. In the current case, the Hong Kong data fit the structure established by the CivEd study, but they can be modelled with the structure established in these studies as well. Although the usual statistical techniques were used in the studies reported here, a considerable emphasis was also placed on the theoretical adequacy of the structure employed. Again, this is not a new procedure, but it sometimes places psychometrics and political theory at odds. The studies reported here are certainly concerned about statistical reliability and validity, but sometimes theory provides a better guide for scale construction. Nevertheless, the interaction between psychometrics and theory is what is important, as demonstrated in the studies reported.
   
   
2. Causal modelling
   Several studies have identified relationships between variables in an attempt to account for the variance in students’ performance (Kennedy, 2010, Kennedy, 2011, Kennedy, Huang & Chow, 2011). In particular, we have moved beyond examining variance in Civic Knowledge scores by investigating Participation as an outcome measure as well. An interesting pattern shows that what supports the development of civic knowledge the most does not always support the capacity for future participation in civic activities. In terms of political trust, for example, trust in socio-legal institutions requires high levels of civic knowledge, but trust in the media and trust in political institutions do not (Kennedy, Huang & Chow, 2011). These results are much more nuanced than those found previously when “political trust” was developed as a unidimensional scale. In terms of expected political participation, students scoring high on Civic Knowledge are less likely to engage in conventional political activities such as voting and are even less likely to engage in protest activities (Kennedy, 2011). This finding is counterintuitive, and it calls for further investigation on the importance of civic knowledge in general. The issue on citizenship education is how best to develop informed and active citizens, but these results suggest that, at least in some cases, being informed is not the key link to participation.
   
   
3. Multiple methodologies
   The Technical Reports (2, 4) show how initial baseline measures are developed using Rasch analysis to complement the original CivEd analyses. In further analyses, either Rasch (Kennedy & Chow, in press) or the traditional test theory (Kennedy, 2010) was used. In a later study (Kennedy, Huang & Chow, 2011), the two methodologies were used to show how they complement each other. Au (in progress) used focus groups to follow up on the administration of CivEd surveys to the 2009 Hong Kong sample. The integration of mixed methods is an experimental feature of this project, and it has yielded interesting results. For example, in Au’s (in progress) study, the focus groups enabled students to explain why they adopt certain citizenship attitudes, which is impossible to derive from survey data alone.

Professor Kerry Kennedy is the Chair Professor of Curriculum Studies and holds concurrent appointments as Dean of the Faculty of Education Studiers and Associate Vice President (Quality Assurance). He is currently Co-director of the Centre for Governance and Citizenship. Prior to joining the Institute, he was the Pro Vice Chancellor (Academic) of the University of Canberra in Australia. He completed his undergraduate studies and initial professional training at The University of New South Wales in Sydney. He has a Master of Education degree from The University of New South Wales and a Master of Letters degree in History from the University of New England. He completed his MA and Ph.D. at Stanford University. While at the Institute, he has won two Public Policy Research grants, three General Research Fund grants, and one Quality Education Fund grant. Prior to coming to Hong Kong, he won two research grants from the Australian Research Council. His research interests are curriculum policy and theory, especially citizenship education. He published Changing Schools for Changing Times - New Directions for the School Curriculum in Hong Kong (Chinese University Press, 2005), which has now been translated to Chinese (解讀香港教育：香港學校課程的新趨勢 - Chinese University Press, 2011). He co-authored Changing Schools in Asia: Schools for the Knowledge Society (Routledge, 2010). Two other successful books he co-authored, Curriculum Construction and Celebrating Student Achievement—Assessment and Reporting (Pearson Education Australia), are now in their fourth editions.