Keith Trigwell

  • A revised ATI for investigating relations between teaching and assessment of student learning

    Keith Trigwell, University of Oxford
    Paul Ginns, University of Sydney

    Research paper

    Themes: learning and teaching methods, assessment

    One of the few available instruments for exploring the relations between the way teachers teach and conceive of assessment, and the ways students of those teachers experience learning, including the assessment demands, is the Approaches to Teaching Inventory. It has now been used in a variety of contexts (disciplines, university types and teaching at different student study levels) and shows, with some consistency, how qualitatively different ways of teaching are related to a range of teaching-related variables. Because of this growing interest and use, in the last few years this instrument has been the focus of critical inquiry by its developers and others. Late last year a detailed description of the development of the instrument was published in a refereed journal (Trigwell & Prosser, 2004). A second, and highly respected, refereed journal has also accepted for publication, a confirmatory factor analysis supporting the validity of the two ATI scales (CCSF and ITTF) (Prosser & Trigwell, accepted). During these investigations it became clear that some of the 16 items were not working as effectively as others, and that in the ITTF scale particularly, the nature of the existing items were thought to be showing a bias against teaching involving quality presentation.

    As a result of these reviews, a revised 25-item version of the ATI was developed and tested in November-December 2004. A total of 318 academic staff from 4 different universities in Australia and one university in the UK completed the new version. Confirmatory factor analyses showed:

    1. that a one factor model is not a good fit with the data (CFI = 0.70, NNFI = 0.68, RMSEA = 0.14, SRMR = 0.12);
    2. that with three items removed, a two-factor solution with one factor containing 11 Conceptual change/student-focused items and a second Information transfer/Teacher-focused factor, also containing 11 items, is a good fit with the data (CFI = 0.95, NNFI = 0.94, RMSEA = 0.06, SRMR = 0.08);
    3. Scale reliabilities of 0.86 and 0.83 for the new CCSF and ITTF scales respectively; and
    4. Factor correlations of –0.36.

    The new ATI-R, which will be made available at the conference, contains 22 items. Both the ITTF and CCSF scales contain 7 of the original 8 ATI items and 4 new items.

    In this presentation, we summarise the origins of the ATI as outlined in the two refereed papers, we present the results of the analysis of the ATI-R, we outline the differences and advantages of the new 22-item version of the ATI and describe some of the ways the instrument can be used in pedagogic research. This last area will be set up as the focus of the discussion as we see this instrument as being an essential part of many investigations involving the ways teachers approach their teaching, relations with their conceptions of teaching, including assessment (Trigwell and Prosser, 1996) and how students experience the assessment environment.


    • Prosser, M and Trigwell, K (accepted). Confirmatory factor analysis of the Approaches to Teaching Inventory, British Journal of Educational Psychology.
    • Trigwell, K and Prosser, M (2004). Development and use of the Approaches to Teaching Inventory, Educational Psychology Review, 16(4), 409-424.
    • Trigwell, K and Prosser, M (1996). Changing Approaches to Teaching: A Relational Perspective, Studies in Higher Education, 21, 275-284.