![](/web/20060126174428im_/http://www11.hrdc-drhc.gc.ca/edd-img/spacer.gif) |
![](/web/20060126174428im_/http://www11.hrdc-drhc.gc.ca/edd-img/spacer.gif) |
10.0 Conclusion
This paper has presented the basics of quasi-experimental evaluation design, with a focus on
considerations for controlling selection bias and for choosing comparison groups. Quasi-experimental designs are usually the best approach to use for evaluating social programs
because one-group designs are inadequate for establishing cause and effect, and two-group
experimental designs are very often impractical. Complex analytical procedures are required to
isolate the effect of the program from many other potential causes under this design,
however.
|