Validity Reflection

The study that I chose to look at for the validity analysis was Kindle, K. (2010). Vocabulary development during read alouds: examining the instructional sequence. Literacy, Teaching and Learning, 14 (1 &2), 65-88. This was a case study involving four primary grade teachers and how they incorporated effective vocabulary practices into read aloud instruction. I chose this study to analzye because it is very similar to my case study.

The author was clear in stating the research questions which were 1) What specific strategies did the teachers use? 2) How did they link into instructional sequences? 3) How did instructional sequences differ among teachers? After a thorough review of the literature, and examples of three research based models of vocabulary discussion, Kindle went on to cleary articulate the methodology employed in this study. The researcher used systematic observation and proves a rationale for doing so as written in “An Observation Survey of Early Literacy Development” Clay (1993). Kindle took the role of non-participant observer during these observations.

While Kindle was transparent in terms of stating clearly her methodology, she does not clearly articulate her role as a research or does not make any of her research biases clear. In addition, she gives a description of the length of teaching of the four participants (two of whom are described as veteran and two are described as new to the profession); yet, she does not explain how the participants were recruited, or selected (i.e. random, purposeful sampling), and while it is implied that these four teachers were chosen because of years of teaching experience, she does not explicitly state this.

The observation process was clearly described and Kindle by stating that she did not give specific instructions to the teacher to avoid leading them to include particular practices. She uses Merriam (2001) to support this technique.
Kindle provides a clear description of the data analysis. She used Creswell (2008) technique of analyzing transcripts in a recursive or spiral manner, used low inference coding (Carspecken, 1996) to establish categories. Transcripts were re-read and patterns of interaction were matched to the model instructional sequences presented. Inter coder reliability was established through the use of peer debriefing, and the author states that member checks were used to increase validity of her depictions of the data. In addition, Kindle describes that she used the data to form a structured interview protocol and the interview helped triangulate the data to further ensure validity. However, while she mentions the topic domains included in the interview, she does not explicitly state some of the research questions, nor does she use any dialogue from the interview in her transcripts. This would have added to the thick descriptions that she made with her transcripts of instructional sequences. Kindle provided vignettes from the transcripts in the analysis and result section to support conclusions and to compare and contrast the teacher behavior to the model instructional sequences.

Finally, while Kindle goes on to state that no data was obtained measuring student word knowledge or word learning because this was a study interested in teacher behavior, she later draws conclusions about effective vs. ineffective teaching practices with only her examples from the transcript to support. It should have been more clearly stated in the results section that these effective/ineffective practices were specific to these cases and warranted further exploration before they could be generalizable.


Overall, this case study was a good model of a case study for my research. The author is effective in establishing validity and reliability. However, to strengthen the validity, there were several areas where the author could have been more explicit. One, researcher bias nor were the limitations to the study implicitly stated. In addition, explicitly stating that the conclusions that she made regarding effective and ineffective practices were based on these specific cases and warranted further exploration before generalizing seems warranted. Finally, while the interviews served to triangulate the data, the interviews were not incorporated into the discussion section and it is unclear how they were used in the study. The author merely provides a description of the topic domains used to develop the protocol. Example of interview questions or language samples obtained from the interviews should have been woven into the discussion/results section.