Abstract: Educational Psychologist Richard Lookatch wrote in the Summer 1995 edition of TECHNOS that research literature that he had monitored since the early 1980’s had never established that using technology improved learning. His analysis was that all of the studies were flawed with Type I errors. Mr. Lookatch’s sweeping generalizations tend to overlook the value of many initial studies into the new field of computer and technology enhanced learning. The author’s dissertation research avoided both Type I and Type II errors.

        A few years ago, Educational Psychologist Richard Lookatch wrote in the Summer 1995 edition of TECHNOS that research literature that he had monitored since the early 1980’s had never established that using technology improved learning. His analysis was that all of the studies were flawed with Type I errors, finding benefits that really were not there, as summarized in the November 1995 edition of Education Digest (Lookatch, 1995).

        As the author was planning his dissertation research in Instructional Technology under Dr. Charles Dickens of Tennessee State University, Nashville, he was made aware of the necessity of avoiding both Type I and Type II errors by several of his professors. Dr. Carol Stice impressed upon her class in Research Methods that researchers can only conclude within a stated probability of error that the decision to reject the null hypothesis is the correct decision. As there is always a possibility that the decision reached by the researcher is wrong, two kinds of incorrect decisions can be made (Crowl, 1996).

        In Type I errors, the researcher incorrectly concludes that the null hypothesis should be rejected, when in actuality there really is a significant difference between the means -- too different to be the result of chance error. In Type II errors, the researcher incorrectly concludes that the null hypothesis should not be rejected, when in actuality there really is not a significant difference between the means (Crowl, 1996; Gay, 1981).

        Type I and Type II errors usually occur when the study is constructed in such a way as to ignore established rules of proper research. Whereas Type I errors involve setting the level of significance in data analysis so low as to find benefits that are not really present, Type II errors involve setting the level of significance in data analysis so high that benefits that are really present are overlooked. Failing to control for a host of conditions that may account for the observed impact on learning also leads to Type I errors (Crowl, 1996).

        The author’s 1998 research study took place in about a dozen 4^th grade classrooms over a period of almost three months. During this time, weekly lesson plans were developed by the author and implemented by the classroom teachers of the six treatment classes. The treatment used social studies simulation software and the Internet, integrated into the regular curriculum in one-computer classrooms. Teacher training was provided and weekly contact was made with each of the treatment classrooms and their teachers.

        Through the guidance of Dr. Charles Dickens, the author’s dissertation research avoided both types of errors. Type II errors were avoided by setting the level of significance at the .05 level, but the actual outcomes of the study showed the data analysis definitely avoided the Type I error as well. The ANCOVA results of the pre-test and post-test data were extremely significant in proving that the author’s treatment could impact critical thinking.

        Mr. Lookatch’s sweeping generalizations that all of the studies that he had reviewed contained a fundamental flaw, i.e. Type I errors, tends to overlook the value of many of these initial studies into this new field of computer and technology enhanced learning. Certainly, the pressure is there for researchers to overstate their case, but scholarly journal review boards and dissertation committees have an obligation to ensure that the research is properly conducted and adequately reported. Such was the fulfillment of the duties of the author’s dissertation committee and of his dissertation chair, Dr. Charles Dickens. The author’s research findings are statistically significant and stand upon their own merits.

References

        Crowl, T. K. (1996). Fundamentals of educational research (2^nd ed.). Madison, WI: Brown & Benchmark.

        Gay, L. R. (1981). Educational research: Competencies for analysis and application (2^nd ed.). Columbus, OH: Charles E. Merrill.

        Lookatch, R. P. (1995). Technology for teaching? The research is flawed. Education Digest, 61, 4-7. [Online]. Available: http://www.ebscohost.com (1998, December 10).

Presented at SITE 99 - Society for Information Technology and Teacher Education International Conference. This annual conference offers opportunities to share ideas and expertise on all topics related to the use of information technology in teacher education and instruction about information technology for all disciplines in preservice, inservice, and graduate teacher education as well as faculty and staff development.

Society for Information Technology and Teacher Education is an international association of individual teacher educators, and affiliated organizations of teacher educators in all disciplines, who are interested in the creation and dissemination of knowledge about the use of information technology in teacher education.