Association of Cognitive Style and Satisfaction With Distance Learning
Nelson A. Foell
The University of Georgia
Robert L. Fritz
The University of Georgia
Universities traditionally offered vocational teacher education courses either on campus or through itinerant educators who traveled to off-campus locations (Roberts, 1957). Over the years, the conditions that affect the delivery of teacher education have changed. Smaller enrollments at a given site, limited monies for travel, and competing demands for time made it necessary for university administrators to consider other teaching options. Recognizing these conditions, faculty in the Department of Occupational Studies at The University of Georgia decided to use new technology to embark upon an alternative distance learning teacher-education delivery program.
While new technology offered the department clear advantages, questions about the learning impact of distance technology remained. Some researchers argued that considerations like this are important when tasks are new, demanding, and/or ambiguous to the learner (Snow, 1989; Witkin & Goodenough, 1981). These tasks can require emotional and cognitive skills that some students have not developed (Snow, 1987; Witkin, Moore, Goodenough, & Cox, 1977). For example, when students are mismatched to task demands, Snow (1989) saw dissatisfaction as a likely outcome. While Fritz (1995) linked such learning mediators as cognitive structure and emotional support to this problem, it has not been clear how this issue might impact distance learning.
This study used theories of cognitive style and attitudes to explore the relationship of cognitive style and satisfaction with distance learning. Field-dependence cognitive style theory was used to explain relevant cognitive and emotional traits (Fritz, 1994; Messick, 1986) and attitudes were studied to estimate satisfaction. A literature review indicated that no previous study of distance learning in vocational education had used this approach.
The Department of Occupational Studies first used distance learning technology to deliver teacher education courses in the Spring of 1986. An electronic bulletin board was used to deliver off-campus instruction to several locations, using a computer and modem as the primary communications mode. A private e-mail mode could also have been used to achieve two-way communication; however, responses were delayed and real-time interactive communication was not possible. Although the system allowed assignments to be made, topics to be discussed, and questions to be answered, the lack of real-time response was a major drawback (Foell & Weitman, 1992). The existing technology at that time could not overcome this limitation.
A new technology became available in 1993 in Georgia--a compressed video system that used telephone lines for the multi-way transmission of voice, pictures, and data. This system was initially established to address medical needs throughout Georgia but was extended to universities, public schools, and post-secondary institutions. Operated as the Georgia Statewide Academic and Medical System (GSAMS), this technology offered a more advanced level of real-time interaction. However, while instructors and students could interact, only two sites could function at any given time.
Consequently, because questions about the effectiveness of this new technology as an instructional vehicle seemed to have implications for instructors and other learning mediators, this study focused on student characteristics. As Lazarus (1982) understood it, the individual is a meaning-maker whose perception is important when tasks are demanding.
The work of numerous researchers is relevant to this study. Sternberg (1987) supported the use of cognitive style theory to examine individual functioning on different tasks. Messick (1987) explored information processing, and Witkin and Goodenough (1981) included attitudinal variables. Witkin and Goodenough viewed these traits as part of a stable personality structure. They also contended that cognitive styles are not widely adaptive; instead, they are effective in a narrow range of challenging situations.
There are situations where cognitive styles do not differentiate performance or attitudes (Snow, 1989; Witkin & Goodenough, 1981). Examples include less complex and demanding tasks (Bloom, 1984) or when instructors provide vital learning mediators (Witkin et al., 1977). These situations do not require students to demonstrate advanced cognitive and emotional resources.
Field-dependence theory. Witkin's cognitive style theory explains factors that are important to this study. That theory contains/recognizes two cognitive styles, field-dependent and field-independent, as opposites on a bi-polar continuum with different information processing and learning style components that suggest simple to complex modes of functioning.
Of the two cognitive styles, the field-dependent is less complex. Witkin and Goodenough (1981) characterized its information processing traits as passive. Because they have not developed sophisticated learning mediators, field-dependent individuals may use a chain-link information processing style. They also accept ideas as presented and do not modify them. They prefer teaching methods that encourage teacher-student interaction and like courses that emphasize social information. Fritz (1994) reported that as many as 70% of secondary vocational students had the field-dependent cognitive style.
In contrast, Witkin and Goodenough (1981) saw the field-independent style as a complex and individualistic cognitive style. Its information processing traits, which include hypothesis-testing and restructuring skill, are examples of cognitive mediators that are used to reason about ambiguous and demanding problems. In addition, their attitudes show that they prefer to learn independently and prefer courses that emphasize abstract and non-personal content, like that found in math and science.
Bertini, Pizzamiglio, and Wapner (1986) note this essential point about cognitive style. By the middle teenage years, a cognitive style is a stable, automatic, and predictable pattern of functioning that is applied across task and sensory situations. They contended, as did Messick (1987), that a stable cognitive style resists modification.
Given this foundation, it appeared that several traits in the new distance learning environment could favor individual functioning. While direct social interaction was mediated by technology, contact on the system was limited because only two locations could interact at one time. Thus, if course content is ambiguous and/or demanding, access to instructional support may be limited. It was not clear that the instructor could provide the support through the distance learning system that is important to some students.
This study used three attitude concepts to explore student perceptions of distance learning experiences. Responses to the concepts were correlated to the field-dependence cognitive styles. Attitudes were then evaluated as independent constructs. The three attitude concepts were
- How the instructor taught the distance learning course,
- the skills and the concepts students learned in the distance learning course, and,
- how students got along with or related to the instructor.
The population for this study consisted of 27 students enrolled in a technology education course offered via distance technologies. Table 1 shows selected population demographics.
Table 1 Student Enrollment by Site, Gender, and Educational Level (N=27)
Variable Sites Site 1
Male 3 6 3 6 Female 3 0 2 2 Graduate 3 5 3 3 Undergraduate 5 1 2 5
The course was titled "Needs Analysis in Technology Education" (ETS 512/712). It was offered for 5 quarter hours during the Spring of 1994, at four different sites (i.e., three remote sites and one on-campus). The instructor had more than seven years experience with distance learning programs. He designed this course to achieve these basic objectives: (1) survey manpower needs using a variety of library and human resources, (2) identify instruments to assess student learning needs, (3) analyze data and sequence instructional objectives, (4) explore the impact of new technology on work place and classroom environments, and (5) identify strategies that portray women and minorities as equals in the workplace.
Instruction involved lectures, computer slide shows, readings, class discussions, student presentations, fax for handouts, and a flat bed graphic stand and visual projector manufactured by the ELMO Corporation (ELMO). The ELMO projector enabled video transmissions to the off-campus sites similar to an overhead projector, but with two major differences: (1) The image is on a television monitor and not a screen, and (2) the material being projected can be a textbook, a multi-dimensional copy, a real object, or a photograph.
The typical class meeting followed this format: Roll call was taken to determine that all remote sites could see and hear and be seen and be heard. The ELMO projector was used to present an illustrated lecture of new material, followed by a discussion period. Students then made presentations based on a topical outline using the fax and ELMO projectors at the remote sites to show diagrams, photographs, or drawings to the other sites. Follow-up questions and answers were then addressed. A quiz might be shown on the ELMO projector and students could fax their answers to the instructor for grading. Feedback could be sent via fax or given orally. Finally, the evening's activities would be summarized and assignments made for the next class meeting.
The course met one evening a week for ten weeks. The students also attended three Saturday sessions on the university campus. The first Saturday session was to review the syllabus and course materials. The second session was for the mid-term exam and the last session was for the final exam, course evaluation, and the submission of required term papers.
The Group Embedded Figures Test (GEFT) and a Semantic Differential Inventory (SDI) were used to collect data from the 27 students. The GEFT measures the field-dependence cognitive styles (Witkin, Oltman, Raskin, & Karp, 1971), while the SDI measures attitudes (Osgood, Suci, & Tannenbaum, 1957) .
The GEFT has 18 simple forms that are embedded in complex or camouflaged backgrounds. Subjects locate and outline the simple forms. Those who have trouble with this task have the lowest scores on the 0 to 18 range; theirs is a field-dependent cognitive style. Field-dependent scores cluster between 0 and 5. Subjects who find the simple forms with greater ease have the highest scores and thus a field-independent cognitive style. Their scores clustered between 13 and 18 on the range. According to Witkin, Oltman, Raskin, and Karp (1971), the GEFT has satisfactory reliability (.89 on test-retest over a three year period) and validity (a correlation of .82 between the two major sub-sections).
While the range of GEFT scores for the study population was from 0 to 17, scores are grouped on Table 2 to illustrate how the field-dependent and field-independent cognitive styles occupy the 0-18 continuum. The GEFT mean was 9.41 or roughly the mid-point, with 68% of all scores (+1 SD) between 15.07 and 3.74.
Table 2 GEFT Scores by Grouping
GEFT Range n M
Field-dependent 0-5 8 2.50 Mixed 6-12 10 9.10 Field-independent 13-18 9 15.89
Note: GEFT Group Data: M = 9.407 SD = 5.667 n = 27
Osgood, Suci, and Tannenbaum (1957) designed the SDI to identify the meanings that people ascribe to specific attitudinal concepts. Three concepts were examined in this study: (a) How [instructor's name] taught in a system designed for distance learning, (b) The skills and the concepts I learned in the distance learning delivery system, and (c) How I got along with or related to the instructor.
After the concepts are identified, Osgood et al. (1957) recommended that researchers use three adjective pairs to measure each of three factors--the activity, evaluation, and potency factors. Activity is concerned with "quickness, excitement, warmth, agitation and the like" (p. 73). Evaluation is concerned with "judgement, [sic] based on rewards and punishment both achieved and anticipated" (p. 72). Potency is concerned with "power and the things associated with it, size, weight, toughness, and the like" (pp. 72-73).
Each adjective pair has positive and negative/high and low ends and is maximally loaded on one factor and minimally on the other two. The adjective pairs used in this study are from factor studies reported by Osgood et al. (1957). Instead of the recommended three adjective pairs, we used four pairs for each factor. These, then, are the 12 adjective pairs that were used in this study, with the positive ends listed first. Activity included complex-simple, fast-slow, active-passive, and excitable-calm. Evaluation consisted of complete-incomplete, meaningful-meaningless, positive-negative, and good-bad. The Potency pairs were soft-hard, lenient-severe, humorous-serious, and light-heavy. The average factor loading was .49, which was satisfactory for this study.
In addition to being randomly placed on the inventory, Osgood et al. (1957) called for a seven-point scale to measure the attitudinal concepts. The positive ends were thus assigned the value of 7 and the negative ends the number 1. By description, the extreme scores (numbers 7 and 1) were labeled extremely positive and extremely negative respectively. Numbers 6 and 2 were labeled quite positive and quite negative. Numbers 5 and 3 were labeled slightly positive and slightly negative. Four was labeled neutral. This distribution allowed good item discrimination and enabled the researchers to determine the direction and magnitude of a score. The scores described in the results section are total scores by factor and concept.
Recommended and standardized test administration procedures were used to collect the data; methods for the GEFT were outlined by Witkin et al. (1971), and the SDI procedures were outlined by Osgood et al. (1957). Data were collected during the final Saturday meeting of the 27 students participating in the study.
Because it is a timed test, students completed the GEFT first. The GEFT consists of three sections, one for practice and two for actual scoring. After the GEFT was completed, students were given instructions to complete the SDI. They also provided basic demographic information. The total time required for data collection was about 45 minutes.
After data were collected, the researchers recorded them on computer scan sheets. Demographic data were collected and used to describe the population. Correlation and descriptive statistics were then used to analyze the data. The data were processed through The University of Georgia Educational Research Services Laboratory.
The results (see Table 3) show no statistically significant relationship between GEFT scores and the SDI factor scores for evaluation, potency, and activity for the first attitudinal concept, How [instructor's name] taught in a system designed for distance learning. There were no statistically significant linear relationships between attitudes toward the instructor's teaching strategies in the distance learning environment and cognitive style.
Table 3 Pearson Correlation for Attitudinal Concept: How [instructor's
name] taught in a system designed for distance learning.
Factor n r M SD p
Evaluation 27 -0.006 5.28 1.453 0.976 Potency 27 -0.213 4.21 1.030 0.286 Activity 27 0.075 4.00 0.636 0.711
Insight to attitudes is gained from the first standard deviations for each SDI factor. While the evaluation factor ranged from 3.83 (neutral) to 6.73 (extremely positive, i.e., complete, meaningful, positive, and good), potency factor scores ranged from 3.18 (slightly negative, i.e., hard, severe, serious, and heavy), to 5.24 (slightly positive, i.e., soft, lenient, humorous, and light).
Activity scores ranged from 3.36 (slightly negative, i.e., simple, slow, passive, and calm), to 4.64 (slightly positive, i.e., complex, fast, active, and excitable). While the widest range in attitudes was for evaluation, activity had the narrowest range. The conclusions and implications of these findings are in the discussion section.
For the second attitudinal concept, Table 4 also shows no statistically significant correlation between GEFT scores and SDI scores for the evaluation, potency, and activity factors for the attitudinal concept, The skills and the concepts I learned in the distance learning delivery system. There were no statistically significant linear relationships between attitudes toward course content and cognitive style.
Table 4 Pearson Correlation for the Attitudinal Concept: The skills and
the concepts I learned in the distance learning delivery system.
Factor n r M SD p
Evaluation 27 -0.043 5.22 1.338 0.832 Potency 27 -0.032 4.30 0.896 0.873 Activity 27 0.064 3.79 0.642 0.750
The first standard deviations on the three SDI factors are similar to the first concept. The range for evaluation was again the largest, from 3.88 (neutral) to 6.56 (extremely positive, i.e., complete, meaningful, positive, and good). And, while the range for potency was from 3.40 to 5.19, scores were modestly skewed toward the positive side (i.e., "slightly" soft, lenient, humorous, and light. Finally, while the range on the activity factor was again the narrowest, from 3.15 to 4.43, scores were modestly skewed toward the negative side (i.e., "slightly" simple, slow, passive, and calm).
With regard to the third attitudinal concept, Table 5 shows that there were again no statistically significant correlations between GEFT and SDI scores on the evaluation, potency, and activity factors for How I got along with or related to the instructor. Again, there were no statistically significant linear relationships between attitudes toward the instructor and cognitive style.
Table 5 Pearson Correlation for Attitudinal Concept: How
I got along with or related to the instructor.
Factor n r M SD p
Evaluation 27 -0.182 5.53 0.986 0.182 Potency 27 -0.063 4.40 0.770 0.755 Activity 27 0.181 4.19 0.591 0.367
With the first standard deviation as the reference, 68% of all scores for the evaluation factor ranged from 4.54 (slightly positive) to 6.51 (extremely positive, i.e., complete, meaningful, positive, and good). They ranged from 3.63 (neutral) to 5.17 (slightly positive) on the potency factor ( i.e., soft, lenient, humorous, and light), and from 3.60 (neutral) to 4.78 (slightly positive) on the activity factor (i.e., complex, fast, active, and excitable). The widest attitudinal variation was again on the evaluation favor, with activity having the more constricted range.
In sum, while there were no statistically significant correlations between GEFT and SDI factor scores, the SDI means and standard deviations suggested levels of satisfaction with distance learning. While the evaluation mean scores were the most positive, they also had the widest range. And, while both the potency and activity mean scores were more neutral and the ranges more restricted, they suggested a more common opinion. However, activity mean scores were the lowest and had narrower ranges.
This study answered two questions about distance learning. The first involved cognitive style and the second involved attitudes. The findings revealed that, while correlations between attitudes and cognitive style were not statistically significant, attitudes toward course content, instructional methods, and the instructor seem to have their primary implications for distance learning in the areas of potency and activity. Osgood et al. (1957) associated potency with perceptions of power and toughness, while activity was associated with excitement and quickness.
The attitudes expressed through the potency and activity factors were different than those for the evaluation factor. While each potency and activity mean score for the three attitudinal concepts was near the neutral point on the range, the three mean scores for the evaluation factor were from "slightly" to "quite" positive. The implications of this finding are discussed below.
In contrast to the definitions for potency and activity, Osgood et al. (1957) defined evaluation as a personal judgment. The findings show that the students rated the teaching methods, course content, and the instructor from slightly to quite complete, meaningful, positive, and good. These findings suggest satisfaction with experiences in the distance learning environment as evidenced by personal judgment.
By comparison, the findings for the potency and activity factors were not strong in their positive or negative orientation. This suggests that students were neither discouraged nor excited with the toughness or the pace of their distance learning experiences. This finding has two implications: One involves an explanation as to why correlations between cognitive style and attitudes were not statistically significant. The second point involves the nature of the distance learning environment.
The findings for the potency factor suggest that the learning experiences in ETS 512/712 were not viewed as demanding and/or ambiguous. Witkin et al. (1977) suggest that this may occur because the instructor mediated the potentially difficult learning with teaching strategies. In general, provisions for learning mediators are more important for field-dependent students than for field-independent students. The way the course was taught, field-dependent students had regular access to such learning mediators as structure and interpersonal contact. The instructor and the students met face-to-face on three Saturdays during the ten week course. There was also provision for discussions and presentations at each session. Each of these methods could provide both the cognitive and psychological support that field-dependent students need when tasks are demanding. However, Witkin et al. (1977) contended that providing them would not impinge learning for students with a field-independent cognitive style.
Cognitive style is not an important factor when there are few problems with learning mediators (Witkin & Goodenough, 1981; Witkin et al., 1977). Snow's (1989) logic argues that learning tasks and methods that require special psychological and emotional skills will differentiate performance. If students had major problems that involve learning mediators, potency factor scores for course content and teaching methods should have been more clearly toward the lower or negative ends of the scales.
The lack of apparent difficulty with the course also seems evident in the activity factor. Among the concepts it represents, Osgood et al. (1957) associated activity with quickness and excitement. While the activity factor mean for how the course was taught was neutral, it was slightly negative for the skills and concepts being learned. Students tended to rate the skills and concepts being learned as "slightly" simple, slow, passive, and calm. This could reflect the pace of the learning experiences.
Distance technology could be viewed as a slower way to learn. Even though GSAMS is an improved distance learning system when compared to previous designs, its real-time nature is still different than traditional learning. No more than two sites can interact at a given time, even though all sites receive the transmission. The system can also break down and a site can have trouble joining the conference loop. There are also times when students want to interact but do not because system constraints limit direct interaction. Thus, while distance learning is more convenient for many students, those in this study did not view their experiences with it as exciting or quick.
While the Department of Occupational Studies will use this information to improve its delivery of distance learning, other educators can benefit from these findings. The most important point may be that instructors need to participate in the broad-based evaluation of the learning effectiveness of their distance learning courses. The insight that is gained can be used to alter the delivery style. This course was not deliberately modified to account for student predilections.
While cognitive style differences were not found in this course, the potential still exists that others could experience difficulty due to differences in student information processing and psychological resources and task demand. While there are clear differences in the technology used to delivery distance learning courses, the most fundamental tenet returns to individual effectiveness in the teaching-learning process. While the instructor must become proficient in the use of video cameras, an ELMO projector, a microcomputer, and a fax machine, the verdict on the value of distance learning technology must ultimately return to its impact on teaching and learning.
Foell and Fritz are Associate Professors, Department of Occupational Studies, College of Education, The University of Georgia, Athens, Georgia.
Bertini, M., Pizzamiglio, L., & Wapner, S. (1986). Epilogue: Relation of Witkin's work to future trends in psychology. In M. Bertini, L. Pizzamiglio, & S. Wapner (Eds.), Field dependence in psychological theory, research, and application (pp. 119-126). Hillsdale, NJ: Erlbaum.
Bloom, B. S. (1984). Taxonomy of educational objectives handbook I: Cognitive domain. New York: David McKay.
Fritz, R. L. (1995, June). The association of field-dependence to preferences for math and science among secondary marketing education students. Research paper presented at the Marketing Education Conclave, Copper Mountain, Colorado.
Messick, S. (1987). Structural relationships across cognition, personality, and style. In R. E. Snow & M. J. Farr (Eds.), Aptitude, learning, and instruction: Vol. 3. Conative and affective process analyses (pp. 35-75). Hillsdale, NJ: Erlbaum.
Messick, S. (1986) Herman Witkin and the meaning of style. In M. Bertini, L. Pizzamiglio, & S. Wapner (Eds.), Field dependence in psychological theory, research, and application (pp. 115-118). Hillsdale, NJ: Erlbaum.
Osgood, C. E., Suci, G. J., & Tannenbaum, P. H. (1957). The measurement of meaning. Urbana: University of Illinois Press.
Roberts, R. W. (1957). Vocational and practical arts education: History, development and principles. New York: Harper & Row.
Snow, R. E. (1987). Aptitude complexes. In R. E. Snow & M. J. Farr (Eds.), Aptitude, learning, and instruction: Vol. 3. Conative and affective process analyses (pp. 1-9). Hillsdale, NJ: Erlbaum.
Sternberg, D. (1987). Intelligence and cognitive style. In R. E. Snow & M. J. Farr (Eds.), Aptitude, learning, and instruction: Vol. 3. Conative and affective process analyses (pp. 1-9). Hillsdale, NJ: Erlbaum.
Witkin, H. A., & Goodenough, D. R. (1981). Cognitive styles: Essence and origins. New York: International Universities.
Witkin, H. A., Moore, C. A., Goodenough, D. R., & Cox, P. W. (1977). Field-dependent and field-independent cognitive styles and their educational implications. Reviews of Educational Research, 47, 1-64.
Witkin, H. A., Oltman, P. K., Raskin, E., & Karp, S. A. (1971). A manual for the group embedded figures test. Palo Alto, CA: Consulting Psychologists Press.
Reference Citation: Foell, N. A. & Fritz, R. L. (1995). Association of cognitive style and satisfaction with distance learning. Journal of Industrial Teacher Education, 33(1), 46-59.