UBC Theses and Dissertations
A study of the variables associated with the acceptance and rejection of A.E.R.C. abstracts Pipke, Ingrid
Adult education is a field of practice which has given rise to an emerging discipline concerned with the creation of its own body of knowledge. The field and the discipline exist in a reciprocal relationship where information is diffused both ways. One method for disseminating information is the Adult Education Research Conference (A.E.R.C.) which promotes research in the discipline and encourages professional collaboration among adult educators. Information dissemination processes are vital to the discipline and field, and are studied through meta-research. In the present study, abstracts submitted to Steering Committees for the Adult Education Research Conference in 1978, 1979, and 1980 were examined to clarify variables associated with acceptance or rejection. The study was grounded in social science literature focusing on variables associated with the acceptance or rejection of manuscripts submitted for publication. A 41-item instrument was developed to assess the characteristics of A.E.R.C. abstracts. As A.E.R.C. abstracts are judged "blind" (i.e., authors are unknown to judges), the study examined "internal" abstract variables. These concerned the content (adult education focus and methodological orientation), the research processes employed, and the composition of the abstract. Procedures aimed at measuring the reliability and validity of the instrument were executed. Expert judges (the 1981 A.E.R.C. Steering Committee) attested to the content validity of the instrument. For test- retest purposes, 97 abstracts were coded twice and 20 were coded three times to yield a mean item stability-across-time coefficient of r=.68. Inter-judge reliability was established by having five judges code nine randomly selected abstracts. A repeated measures analysis of variance showed that the five judges made consistent decisions concerning 37 of the 39 variables. During a second procedure, the coding decisions of the researcher were compared with those of the judges. "Researcher-judges" data were subject to analysis of variance which revealed acceptable levels of agreement on 37 variables; the two "unreliable" results stemmed from the non-conforming decisions of a judge, not the researcher. During pilot procedures, scales and coding criteria were systematically refined. It was concluded that the final form of the instrument was content valid and reliable. Using this instrument, 329 accepted and rejected A.E.R.C. abstracts were coded on 39 variables. Item means of abstracts accepted and rejected in 1978, 1979, and 1980 differed significantly on nine, six, and nine variables respectively. Variables differentiating between accepted and rejected abstracts were entered into discriminant function equations for 1978, 1979, and 1980. Profiles for accepted abstracts differed by year. In 1978, accepted abstracts were primarily written in an active voice, had a clear and logical argument, were oriented towards use of a particular research methodology, had "clearly identified" instrumentation and implications for the field, and did not focus on agency sponsorship of adult education programmes. In 1979, accepted abstracts were methodologically oriented, focused on programme planning issues but not agencies, had a clearly defined inductive theoretical development, and were not well anchored in the literature. The 1980 "profile" showed that accepted abstracts focused on foundations of adult education or characteristics of adults and learning, had "clearly identified" data collection procedures, used higher-order (e.g., multivariate) data analysis, and only moderate amounts of dysfunctional jargon. Separate discriminant function equations for each year successfully classified 81 percent of abstracts in 1978, 71 percent in 1979, and 78 percent in 1980. It was significant that, in general, variables associated with acceptance did not have the same, or even a similar, effect in each of the years studied. Judges appeared to weight variables differently by year. This raises questions concerning the abstract selection process and the election of Steering Committee members.
Item Citations and Data