Limited research findings have suggested that field-independence-dependence (FID), a dimension of cognitive style, may be a biasing factor in student ratings of instruction. This study experimentally investigated the extent to which field-dependent (FD) and field-independent (FI) students differentially perceived a lecture presentation in which (a) lecture content coverage and (b) instructor expressiveness (enthusiasm) were systematically manipulated. The four dependent variables were student scores representing: (a) a global rating of overall presentation, (b) specific ratings of lecture content, (c) specific ratings of instructor expressiveness, and (d) student achievement.The four treatment conditions consisted of 20-minute videotaped lectures that differed systematically on lecture content (high, low) and instructor expressiveness (high, low). These lectures were presented by a professional actor who spoke on the topic "The Biochemistry of Memory." The high-content lectures purportedly contained 19 teaching points, and the low-content lectures, 2 teaching; points. These and two other medium-content lecture videotapes have been used in previous research to examine the Dr. Fox effect (the tendency to give hi,-;h ratings to an enthusiastic lecturer irrespective of the number of teaching points presented).The subjects were 120 female students enrolled in 12 sections of an undergraduate educational psychology course. Approximately 70% were education majors. Using the Group Embedded Figures Test, one-half of the subjects was classified as field dependent (FD), and the other half, field independent (FI).The subjects were randomly assigned to view one of four videotaped lectures which were shown simultaneously in four classrooms on seven different occasions during a regularly scheduled class period. They were told that an instructor at another university was in the process of developing a series of instructional videotapes and was interested in receiving student feedback in order to make necessary improvements. After viewing the lecture, students completed a rating form and a 19-item achievement test.This study employed a 2 x 2 x 2 (Lecture Content x Instructor Expressiveness x FID) three-way balanced factorial design. All factors were considered fixed. Alphas were set at the .05 level. Comparisons made a priori between FD and FI rating scores were tested using the F statistic. No significant differences were found between FD and FI student ratings of instruction; however, all observed differences were in the directions hypothesized. Multivariate and univariate analyses of variance were completed for the overall-presentation, lecture-content, and instructor-expressiveness subscale dependent variables. Student scores on each of the three rating subscales were significantly higher for the high-expressive instructor than for the low-expressive instructor. Also, student scores on two of the rating subscales (overall presentation and lecture content) were significantly higher for the high-content lectures than for the low-content lectures.A separate analysis of covariance, treating Scholastic Aptitude Test-Verbal scores as the covariate, was used to test hypotheses for student achievement scores. Achievement was significantly higher for the high-content lectures than for the low-content lecturessignificant interaction between lecture content and student FID was interpreted as follows: (a) FI groups outperformed the FD groups for the high-content lecture conditions and (b) FD and FI groups did not differ for low-content conditions. The manipulation of instructor expressiveness did not affect student achievement.The findings did not support the Dr. Fox effect. They are: (a) student FID characteristics appeared to have no effect on student ratings; (b) large differences in the amounts of informational material presented significantly influenced student responses both to a global rating item and to specific rating items designed to measure lecture content; (c) extremely expressive instructor behaviors positively influenced student ratings regardless of the intended measurement purposes of the rating subscales; (d) when specific rating items were used, student ratings were more sensitive to extremes in instructor expressiveness than to extremes in lecture content coverage.
Identifer | oai:union.ndltd.org:BSU/oai:cardinalscholar.bsu.edu:handle/180184 |
Date | 03 June 2011 |
Creators | Roeser, Thomas Dale |
Contributors | Miller, Ebert L. |
Source Sets | Ball State University |
Detected Language | English |
Format | vii, 132 leaves : ill. ; 28 cm. |
Source | Virtual Press |
Page generated in 0.001 seconds