February 17, 2001
Reference: Fuller, H. (2000) First Teach Their Teachers: Technology Support and Computer
Principle: In First Teach Their Teachers: Technology Support and Computer Use In Academic Subjects by Hester Fuller, presented in the Summer 2000 issue of The Journal of Research on Computing in Education, the author examines the following research questions:
Type of Design: The author attempted to portray this as an experimental design but I saw no evidence of a control group. Instead data is compared to national "norms" although the author freely admits that at least in terms of SES, this is not a normal population.
Her methodology for examining the research questions involved the utilization of existing data gathered from the second implementation of the IEA CompEd Study with a sample size of 3,805 (n=3,805) students in 167 grade 5 and 11 classrooms. Roughly 4% of classes (n=225) were set aside for lack of information or identifying characteristic. This sample consisted of a roughly even number of males and females with a preponderance of white (70%) students. From the SES indicators available this sample was slightly more affluent than the national average. The author regressed the above utilizations of coordinator time as the independent variable while maintaining average minimum student use as the dependent variable. The results of the study indicated that by utilizing a fitted models approach to the data analysis there was a significant impact upon the dependent variable p=.04 for grade 5 only in response to question one and a significant impact upon the dependent variables in question two in terms of writing lesson plans with a p=.02 at the 11th grade level.
Factors Jeopardizing Internal Validity: No information was given about the instruments reliability or validity so it was difficult to determine internal validity. The presentation of information gave the impression of a "fishing expedition" which raised serious questions in my mind regarding validity.
Factors Jeopardizing External Validity: The group examined, while having a large sample size was not normed in terms of SES and other key indicators limiting the usefulness of this study.
Adequacy of Statistical Procedures Used: Multiple regressions were used as the primary means of obtaining the data with some type of modeling (which I did not understand) being the means by which conclusions were drawn.
Brief Summary of Logic (Inductive/Deductive): Certain aspects of a technology coordinators job might be more impacting upon student achievement than others. Note that she apparently assumes that having a technology coordinator will impact. This is something that I feel should not be a given.
Design Improvement: Simplify it and/or break it into several different studies with true populations that are generalizable. Provide adequate information about the instrument so we are not left hanging as to internal validity.
Extension of the Study: This study is already over extended. I think she should have started first with the question. Does having a technology coordinator impact instruction? From there the idea could have been narrowed etc.