Quick Guide to Evaluation Approaches

Which type of evaluation approach should I use?

There are a number of different types of evaluation, the most widely used are formative, process and summative (outcome) evaluation. The following information provides you with a guide to the types pf evaluation, when they are used and the types of approaches and data collection methods that are associated with them. (Roberts, Cavill and Rutter 2009)

Formative	A type of process evaluation that assesses how the intervention / programme can be improved as it is being implemented Takes place before the project starts and throughout pilot phase to inform implementation. Shows what worked well and why, and identify challenges to capture lessons learned (continually improving) Utilises evaluation theory e.g. Theory-of-Change; Realist Evaluation, Experience-Based Design, Critical Path etc. Helps to assess needs; clarify theory-of-change; to identify areas for shared learning and improvement Qualitative methods: Literature Review; Interviews; Focus Groups; Observation; Diaries; Questionnaires Service Improvement methods: Stakeholder Analysis; Logic Modelling; PDSA Cycles; Rapid Evaluation Cycles
Process	Assesses implementation and delivery of the initiative, and whether this was carried out as planned Takes place throughout the project implementation and delivery phase. Demonstrates to stakeholders that project objectives are being met Qualitative methods: Interviews; Focus Groups; Questionnaires Quantitative methods: Metrics, Benchmarking, quantifiable Performance, Financial, Service Usage data Service Improvement methods: PDSA Cycles; Rapid Evaluation Cycles
Summative	Was the project successful? Has the met its predefined aims and objectives? Looks at the outcomes (i.e. measures the changes that have occurred as a result of the programme) and/or impact (i.e. the longer, deeper, systemic changes). Demonstrates to stakeholders if project aims were met. Qualitative methods: Interviews; Focus Groups; Questionnaires Quantitative methods: Metrics, Benchmarking, quantifiable Performance, Financial, Service Usage data Can be Experimental (e.g. randomised controlled trial), Quasi-Experimental e.g. comparing before and after or use service Improvement methods: PDSA Cycles; Rapid Evaluation Cycles

Visit the Institute for Healthcare Improvement for more information.