An Introduction to Evaluation Part I: The Formative/Summative Dualism

Scriven’s (1980) formative and summative dichotomy is arguably the most accepted and renowned evaluation typology. One could suppose that the formative/summative dichotomy is therefore fundamental to understanding evaluation. In this post, the concepts of formative and summative evaluation will be explored.

The label, formative evaluation, refers to an evaluation approach that focuses on evaluation improving a programme, essentially allowing an evaluator an interactive (rather than independent) role (Herman et al, 1987). Reporting occurs throughout an evaluation (rather than at the end or at a certain point) meaning closer and sustained contact with a programme (Clarke, 1999). Summative evaluation focuses on formally reporting findings at a certain point in time.

Prescott et al (2002) emphasise that the two approaches are separate and distinct processes. According to Patton (1996), the formative/summative dichotomy, ‘captures the entire array of evaluation purposes’, as it suggests that, ‘anything that is not formative is summative’.

There is much criticism of the summative/formative dichotomy in modern evaluation: ‘the world of evaluation has grown larger than the boundaries of formative and summative evaluation’ (Patton, 1996). Debate exists over which of these approaches is most vital, Cronbach (1966) suggests that formative evaluation is more important that summative evaluation, whereas Scriven (1967) notes strengths of the summative approach. Patton (1996) also suggests that, ‘formative evaluation rests in the shadow of summative’.

At the simplest level, the dichotomy offers a labelling system to categorise evaluation. Indeed, various labels are applied to evaluation (Scriven’s dichotomy included).  McKie (2003) finds that evaluation labels and typologies are excluding to those who are not familiar with the terms (for example stakeholders within the programme being evaluated). Further, Ussher and Earl (2010) suggest that the terms summative and formative can be confusing, however, they do note some value in applying labels: ‘the identification, definition and consistent use of specific labels are useful for developing understanding and communicating with others’.

Where evaluation is dual-level, that is, an evaluation is required at both national and local levels (Allen and Black, 2006), both summative and formative types could still be utilised. A formative evaluation approach could be applied to local-level evaluation with the increased likelihood of the evaluation being used to generate improvements (according to Herman et al, 1987). A summative evaluation type might occur at a national-level evaluation with the likely audience being policymakers, funders and the public according to Herman et al (1987).

This post has explored formative and summative evaluation (in the context of public programmes). I will save some of the more specific debates of evaluation’s purpose , for instance, of evaluator independence (should an evaluator be assisting a programme to improve through formative evaluation? etc.) for another post.

Part II of the Introduction to Evaluation series (available soon) will consider the history of evaluation.