For 2012 the OECD has commissioned two large scale surveys in the education sector; the fifth iteration of the Programme of International Student Assessment (PISA) and the second iteration of the Teaching and Learning International Survey (TALIS). In 2012 PISA has a main domain focus of Mathematics whilst TALIS builds on the scales developed in the in 2007/8. Australia has undertaken to be a part of the TALIS_PISA link study and to interrogate the data to identify relationships between the major domains investigated in TALIS with the student outcomes in PISA.

This paper proposes to report a preliminary investigation of the field trial data of PISA combined with the TALIS data from the same pool of schools. It proposes exploratory analyses of the data through correlation, ANOVA and MANOVA, and multi-level modelling techniques to identify plausible relationships and explained variation that may be uncovered within the data. This investigation will inform the types of analyses that may be performed on the main study data that are being collected in mid to late 2012.

The first iteration of TALIS was implemented across 24 countries at the ISCED 2 (lower secondary school)level focusing on the global domains of:

Conditions for effective learning;
Preparing and supporting a high-quality teaching force;

Improving teaching practice;
Supporting effective teaching through appraisal and feedback; and

Shaping the development of teachers through effective school leadership.

 TALIS sought to provide a validation and broadening of the previous research and to provide a robust design to develop scales and provide a basis for the development of international comparisons and national baseline data.

 The TALIS Board of Participating Countries is extending the scope of the program to include survey data from ISCED 1 (Primary school teachers) and ISCED 3 (Upper Secondary schools teachers) for countries on a voluntary basis. This strategy resulted in an overlap of the populations of PISA and TALIS (ISCED 2 and ISCED 3 students). In order to maximise the utility of the data from both programs it was further decided to offer a PISA_TALIS link program that specifically targeted PISA participating schools to also participate in TALIS. Mathematics teachers will be over-sampled to provide sufficient curriculum centred cases to draw inferences about teachers and student performance.

 The analyses are at a global level by state. Data from PISA participating students are confidential but may be aggregated at school level. TALIS data are confidential at school level with only global estimates at state level being possible. The strict confidentiality protocols for both programs negate any possibility of a one-to-one student to teacher relationship being identified or implemented.

 The proposed methodology is to initially replicate the analysis techniques for each program with the trial data and then to perform supplementary data mining and analyses in an attempt to find relationships that may be further explored in the main study that involves 150 PISA TALIS link schools and surveys a minimum of 6000 students and 3000 teachers.

• Kolen, M.J. (1999). Equating of tests in Masters,G.N and Keeves, J.P. Advances in Measurement in Educational Research and Assessment • OECD (2010). Creating Effective Teaching and Learning Environments: First Results from TALIS http://www.oecd.org/edu/talis/firstresults • OECD (2010). TALIS 2008 Technical Report. http://www.oecd.org/dataoecd/16/14/44978960.pdf • OECD (2011) PISA 2009 Technical Report (Preliminary version) • Wu, M. L., Adams, R. J., Wilson, M. R., Haldane, S.A. (2007). ACER ConQuest Version 2: Generalised item response modelling software [computer program]. Camberwell: Australian Council for Educational Research.