Friday, April 1, 2011

Validity Issues: Wanzek & Vaughn, 2009

While rereading the Maxwell (2005), Chapter 6, “Validity,” what seemed to connect most with a study I am using for my mini-inquiry lit review (Wanzek & Vaughn, 2009; cited below) was the concept of generalization in qualitative research. As Maxell points out, “external generalizability is often not a crucial issue for qualitative studies. Indeed …a qualitative study… may provide an account of a setting or population that is illuminating as an extreme case or ‘ideal type’” (p. 115). Wanzek & Vaughn examined case studies of three students who exhibited extremely low response to reading intervention. More intensive forms of intervention were attempted on the subjects, and it was demonstrated that one of the subjects responded to a more intensified version of the interventions previously tried, while another subject again failed to response to the more intense intervention, and required a whole new approach (whole word reading) to begin responding to the interventions (the third subject dropped out of the study). Clearly, Wanzek and Vaughn’s subjects were extreme cases, and the intense intervention strategies which finally worked on the case study subjects would not be needed for the vast majority of students, so the external generalizability of the study results would be limited. However, as Waznek & Vaughn point out, “teachers of students with disabilities will encounter students with pervasive disabilities in reading and benefit from documented case studies of similar students” (p. 161). As is common to much qualitative research, especially case studies, even though broad external generalizability is not to be expected, studies of extreme cases can still be valuable. The results of these case studies are needed for small populations, for example for students for whom traditional teaching methods and methods developed through large quantitative studies have failed. For this reason, studies like Wanzek and Vaughn, though their application is limited, are still worthwhile. Another validity issue discussed in Wanzek and Vaughn (2009): fidelity of implementation of the interventions. The study states that “six tutors …were hired and trained by the research team…” (p. 154). The tutors received 15 hours of training in instructional techniques, lesson planning, progress monitoring, and group management techniques. The researchers reviewed the tutors’ lesson plans and ran practice sessions with them. Throughout the interventions, the tutors were observed each week and given feedback on their implementation, and monthly fidelity checklists were completed for each tutor. Each tutor received a rating for the quality of their implementation, and ratings for all tutors were consistently high. Knowing that the tutors were trained and monitored so well gave me confidence that the interventions were implemented with fidelity, and consistently for each subject. Wanzek, J., & Vaughn, S. (2009). Students demonstrating persistent low response to reading intervention: three case studies. Learning Disabilities Research & Practice, 24(3), 151-163.

4 comments:

  1. I clearly see the need to do such studies because even those students who do not match the broader and larger studies still need help and still need to be recognized. The issue is common in qualitative research, how do the researchers maintain their stance and credibility when studying people or topics that can be so individualized and situation-specific. As in the case I read, I realize the importance and the need for justification in qualitative research. Sometimes it seems like the researcher has to prove his or herself worthy of being called a true researcher

    ReplyDelete
  2. Looking at the post again, I apologize for the formatting. When I typed it in Word, it was divided into 3 paragraphs, and the citation was properly formatted. Haven't figured out why that happens yet...

    ReplyDelete
  3. @Margaret, it's a problem in the move from .doc to html formatting. You actually have to add paragraph markers to the html page. They don't translate, I think. This has been my experience on my personal blog.

    Nice analysis of the article. It's important to train anyone involved in any study. With qualitative research that can mean data collection agents, instructional deliverers (in the case of an intervention study), coders, etc. It's very, very difficult to do quality research in isolation, and all members of the team need to be on the same page. However, it's also good to have divergent perspectives, which is another reason why teaming in qualitative research is a good thing.

    ReplyDelete
  4. Hmm...sounds like using the team approach in qualitative research adds richness, but must be handled with care so that we understand what our outcomes are really saying.

    ReplyDelete