Lecture 7 (October 11th)

 The Action Research Cycle:

ARcycle.png

t

Diagnosing

Action Planning

Specifying Learning

Evaluating

Action Taking

 

Baskerville and Wood-Harper say that AR has three distinctive characteristics:

 

Five AR evaluation criteria from Davidson et al, (2004): ”Principles of canonical action research”, Information Systems Journal, vol. 14, pp. 65-86.

 

”Action research allows a researcher to test a working hypothesis about a phenomenon of interest by implementing and assessing change in a real-world setting. By analyzing discrepancies between the hypothesized and actual changes [..] the researcher gains both practical and theoretical knowledge about the phenomenon.” Lindgren et al. (2004): Competence Management Systems, MISQ 28 (3),

 
Checkland and Holwell:
Research is ”organized use of rational thought”.There are three domains involved:
Theoretical/conceptual ideas: Framework (F)
 Methodology (M)
 A
rea of interest (substantive/empirical domain) (A)

We can learn about all of these, i.e. they are all susceptible to change. Therefore: Your stand on all three must be established and clarified prior to the study for change/learning to be discovered. To ensure quality in AR research - The recoverability criterion: In AR we cannot achieve ”replicability” (natural science ideals), but should aim higher than ”plausibility”.  The process should be recoverable: you should make clear to interested observers the thought processes and models which enabled the team to make their interpretations and draw their conclusions.

Kalleberg’s emphasis on quality:

Quality in 1) documentation and 2) argumentation

One example project (Caroline Ngoma's Action Research project with HISP Zanzibar) was presented to illustrate how these criteria can be met.