No abstract

No abstract

No abstract

No abstract

Experimenters who attempt to gain and maintain control over research situations routinely apply research designs using control groups. In this article, the authors examine the data produced by an action research project. Their findings indicate that the use of control groups can lead to incomplete and erroneous data and can even force termination of the experiment. The authors offer some suggestions for avoiding these difficulties and outline a new, more appropriate paradigm for research evaluation.