Session 6 (March 21 – April 3)
Large-scale evaluation studies can answer causal questions about program impact. Well-designed studies can learn from implementation about when and for whom programs work. In this session, Bill will interview researchers about their approach to evaluating selected activities of the National Writing Project.
- Many clients want to know the impact of their programs, and experimental designs provide the best evidence of impact.
- Experimental designs are possible to carry out at relatively low cost when: programs are oversubscribed or conditions allow for a delayed treatment, good outcome measures are available and do not have to be collected by the evaluator.
- Prior evaluations of similar evaluations provide important clues as to likely outcomes, challenges, and unintended consequences of programs and policies.
- Evaluation analyses typically require synthesis across different forms of data that do not all point in the same direction.
Post to forum: What applications are there of this evaluation to connected learning evaluations? What would be hard to replicate?