Learning to reason: the factorial survey as a teaching tool in social work education

Abstract

This article proceeds from the notion that the ability to reason well – i.e. to try to find justified answers to questions such as ‘what is the case?’ and ‘what ought to be done?’ – constitutes an essential professional skill, which can be acquired and advanced in professional education. Accordingly, drawing on the much-used argument model developed by the philosopher Stephen Toulmin, we present a conceptual model for professional reasoning and propose an innovative tool for exercising reasoning in professional education. The tool involves the sociologist Peter Rossi’s factorial survey method, which is based on the use of experimentally constructed fictive descriptions (vignettes). By having students make judgements of a large number of short vignettes (N = 96) and conducting statistical analyses separately for each individual, we construct judgement profiles displaying the causal predictors of each student’s judgements. These profiles are subsequently employed in reflective discussions and for constructing arguments in accordance with the Toulmin model. By focussing in detail on the structure and contents of the arguments underlying each student’s judgements, this teaching tool provides a number of advantages. For example, it can be used for making arguments relying on less reliable knowledge sources more transparent and thus open to scrutiny. Further, due to its experimental features, the tool makes it possible to analyse results of both reflective and intuitive thinking. Furthermore, it offers a way of encouraging the students to deal in an analytic and deliberate manner with the potential cognitive errors resulting from intuitive thinking, such as stereotyping.

Keywords:

• Reasoning
• professional judgement
• factorial survey
• vignettes
• Toulmin model

Acknowledgements

The authors would like to thank Professor Hermann Dülmer, Julius-Maximilians University of Wuerzburg (Germany), who generously helped us with drawing the D-efficient vignette sample that was used in the study.

Notes

1. This is John Rawls’ term for factors that may create disagreement among reasonable persons. He lists six sources of reasonable disagreement (Rawls 1993 Lecture II, § 2): (1) Relevant facts in a case can be complex, contradictory and difficult to assess because they point in different directions. (2) Even if we agree about which considerations are relevant in a case, we can disagree about their weight and therefore arrive at different conclusions. (3) To a certain degree all our concepts are indeterminate and vulnerable to hard cases. The use of concepts must therefore be based on judgements and interpretation, in the context of which reasonable persons can disagree. (4) The experiences we have had during our lives shape how we select facts and how we weigh moral and political values. In modern societies with many different positions, a large number of different ethnic and social groups and many kinds of division of labour, people’s experiences are different enough to make assessments different, at least in cases characterised by some level of complexity. (5) Most often there are normative considerations on all sides of a case that vary in their force, and an overall assessment of these considerations can be difficult. (6) One cannot realise all possible positive values simultaneously. This means that one must rank values which can be equally good per se. In the context of such rankings we usually lack clear and uncontroversial criteria.

2. In addition, when the data from previous factorial survey studies in the area of professional judgements were explored for the possible existence of interaction effects, very few such effects were identified (e.g. Wallander and Blomqvist 2005b, 2008; Samuelsson and Wallander 2014, 2015).