Two Views from the Bridge: A Comparison of the Perceptions of Students and Instructors of Elements in the Audio-Teleconferencing Environment |
VOL. 8, No. 2, 1-17
This paper describes an investigation of student perceptions of the audio-teleconferencing environment and a comparison of those perceptions with those of a group of distance instructors. This research is part of an ongoing program that investigates an instructional model proposed by Kirby and Boak (1987), suggesting that elements in the audio-teleconferencing environment can interact with instructor predispositions to determine instructional strategies.
A Q-Sort and a brief questionnaire were used in this study. The Q-Sort, which was constructed for a previous study of audio-teleconferencing instructors, was composed of 79 items. It was used to ascertain the relative importance students attach to factors in the audio-teleconferencing environment. The questionnaire was composed of Likert-type statements and investigated perceptions and attitudes concerning audio-teleconferencing instruction.
The results indicate that the students attach the greatest importance to factors in the instructional environment that are closely related to the quality of the learning transaction. Cluster analysis of the student data reveals two types of students: those who attach more importance to student characteristics and who are thus labelled student-centred; and those who attach more importance to the instructional act and who are, therefore, typified as instruction-centred.
The comparison of the perceptions of students and instructors reveals a number of differences. The students attach more importance to student characteristics and factors relating to the availability of courses, while the instructors rank instructor characteristics and the more abstruse elements, such as the goals of education, more highly than the students.
Le présent article décrit une enquête menée auprès d'étudiants sur leur perception de certains éléments propres à l'environnement de l'audioconférence et compare les résultats à ceux d'une enquête antérieure effectuée auprès d'un groupe de téléformateurs. Cette recherche s'inscrit dans le cadre d'un programme d'investigation d'un modèle pédagogique proposé en 1987 par Kirby et Boak qui suggère que certains de ces éléments, interagissant avec les prédispositions des instructeurs, déterminent les stratégies pédagogiques.
Un Q-Sort, conçu pour l'enquête auprès des téléformateurs, a été jumelé à un bref questionnaire. Le Q-Sort, composé de 79 éléments, visait à identifier l'importance relative accordée par les étudiants à des facteurs de l'environnement de l'audioconférence. Le questionnaire prend la forme d'une échelle de Likert et vise à préciser certaines attitudes et perceptions concernant la formation par audioconférence.
Les résultats indiquent que les étudiants accordent le plus d'importance aux facteurs liés de plus près à la qualité de la transaction d'apprentissage. L'analyse typologique des données-étudiant révèle l'existence de deux types d'étudiants : ceux qui attachent plus d'importance aux caractristiques des étudiants, et ceux qui attachent le plus d'importance à l'acte pédagogique.
La comparaison des perceptions des étudiants avec celles des formateurs révèle un certain nombre de différences. Les étudiants attachent davantage d'importance aux caractéristiques des étudiants et à des facteurs relatifs à la disponibilité des cours alors que pour les formateurs, les caractéristiques de l'instructeur et des éléments plus abstrus tels les objectifs de l'éducation sont signalés comme étant plus importants.
Despite the advent of newer and more exotic technologies, audio-teleconferencing remains a commonly used means of delivering education to post-secondary students at remote sites. The popularity of this medium derives mainly from its relatively economical operating costs and the inter-active capability that it affords to the distance educator. While the interactive nature of audio-teleconferencing makes it similar in some respects to face-to-face instruction, even to the extent that it can be viewed as part of the mainstream educational field (Garrison and Shale, 1990), there are important differences that characterize it as a distinct instructional system. A model proposed by Kirby and Boak (1987, 1989) has served as a basis to investigate this system. The model proposes that elements in the audio-teleconferencing environment, such as instructor characteristics, student characteristics, technical characteristics, and so forth, interact with instructors' predispositions to, in part, determine instructional strategies.
A previous paper by the authors (Kirby and Chugh, 1992) investigated the audio-teleconferencing instructors' perceptions of elements in the instructional environment. That research indicated that the instructors perceived their own characteristics relating directly to the teaching process, such as aptitude for teaching and verbal skills, as the salient elements in the instructional environment. Interestingly, the research also revealed two clusters of instructors, which, it was suggested, paralleled two paradigms of distance education (Holmberg, 1990).
The research reported in this paper focuses on the perceptions held by students of some 79 elements in the audio-teleconferencing instructional environment and the comparison of those perceptions with the perceptions of the instructors reported previously.
In order to investigate the perceptions of the students, the researchers used Q-methodology via the administration of a Q-Sort. The Q-Sort, originated by Stephenson (1953), is a technique for rank ordering items and statistically analysing the results. This technique enables investigators to investigate subjective phenomena, such as attitudes, beliefs, or values, in real-life situations where it is not practical to control factors the way it might be possible in an experimental situation.
It is unrealistic to expect that students will attach the same importance as instructors to different elements in the instructional environment and vice versa. The two groups differ in motivation, circumstances, and experience and are, in all probability, limited in their knowledge of some aspects of each other's personal environments, such as the home circumstances of the students and the institutional environment of the instructors. A knowledge of these perceptions, however, could be of use to both the instructors and the students as well as to the administrators of distance education units. At the very least, this knowledge might help instructors and students to understand each other's values, motivations, and possible behaviours, and for the administrator who is faced with the pragmatic problems of supporting both partners in the learning transaction, the information will be useful for the design of policies and systems.
The Q-Sort instrument used in this study was an instrument constructed for the investigation of the perceptions of distance education instructors (Kirby & Chugh, 1992). The 79-item instrument had been constructed from a pool of items in the following categories: instructor characteristics, student characteristics, administrative factors, technological factors, and the goals and aims of education. The resulting instrument had a repeat reliability co-efficient (r) of 0.74 with a S.D. of 0.16.
The procedure used for the administration of the Q-Sort required the students to sort the 79 items, each of which was printed on a card, into groups along a continuum in response to the following criterion statement:
You are planning to take a course and/or program in Winter, 1991. You are asked to rate the enclosed 79 factors from least important to most important for the success of the course and/or program.
The number of items to be placed in each group was pre-determined in order to obtain a quasi-normal distribution of the 79 items along the 9-point continuum as shown below:
In addition to the Q-Sort instrument, a short questionnaire was constructed for the purpose of this study. The questionnaire entitled, "Experience with AudioTeleconferencing," was composed of ten 5-point Likert scale statements or items designed to explore some of the issues raised in the previous study (Kirby & Chugh, 1992). The questionnaire examined various elements of teleconferencing as well as the individual's experience with teleconferencing.
The Q-Sort was administered to students enrolled in distance education courses at the University of Calgary. The 169 students enrolled in the Fall of 1991 audio-teleconferencing program at the University of Calgary were contacted by the authors, and initially 131 agreed to participate. Only 88 students, however, completed and returned the Q-Sort. In addition, the short questionnaire was administered to these distance students and also to the distance instructors who took part in the previous study (Kirby and Chugh, 1992). Out of the 87 instructors who took part in the initial administration of the Q-Sort, 36 completed and returned the questionnaire.
The data from both the Q-Sort and short questionnaire were assigned numerical scores for the sake of multivariate and bivariate analysis. For the Q-Sort responses the scores ranged from 1, least important, to 9, most important. For the short questionnaire responses, the scores ranged from 1, strongly disagree, to 5, strongly agree.
The analyses conducted were Ward's method of cluster analysis to investigate the existence of homogeneous sub-groups (Ward, 1963), multivariate analysis of variance (MANOVA) to investigate overall group differences, and one-way analysis of variance (ANOVA) to obtain differences resulting from individual variables (Bray and Maxwell, 1985). A statistically significant MANOVA indicates, through statistically significant Wilks Lambda and Multivariable F, the likely presence of group differences. Univariate F's, on the other hand, provide the list of significant individual variables behind the group differences. Only the students' data from the Q-Sort was subjected to cluster analysis for this study. See Kirby and Chugh (1992) for the results of cluster analysis on the instructors' data set. When the sample size was not large enough or group sizes prevented the use of MANOVA, one-way analysis of variance was conducted to investigate group differences. MANOVA was performed on 79 Q-Sort items to determine the overall differences between students (n=88) and instructors (n=87), this analysis was followed by ANOVA to provide the list of items contributing to the overall group differences. ANOVA, without MANOVA, was performed on 79 items on two student clusters with sample size of 29 and 59 and on the 10 variables measured by the brief questionnaire to determine differences between students (n=88) and instructors (n=36).
MANOVA and ANOVA were all conducted by means of SPSSX (Statistical package for the Social Sciences), and cluster analyses were performed using the Cluster Software package (Wishart, 1987).
Table I shows the mean rankings of the 79 items resulting from the student Q-Sorts. The mean rankings ranged from 1.80 to 7.50 on the 9 point scale (1 = least important to 9 = most important). Of the six items whose mean rankings were equal or greater than 7.0, three were directly related to the student - personal goals, motivation, and ability and aptitude. Three were instructor characteristics - aptitude for teaching, verbal skills, and knowledge of the structure of the curriculum. The top ten ranked items included four student characteristics, four instructor characteristics, one - the quality of sound - technical characteristic and one administrative factor - library support services. Virtually all of these elements are closely related to the quality of the learning transaction rather than the more abstract elements, such as the goals of education, which are more peripheral to the mechanics of instruction.
Some items, which are often raised by instructors as being determinants of the quality of instruction, were perceived by the students as being of somewhat lesser importance. Thus, class size, 4.72, and group size at the site, 4.31, received mean rankings close to the mid-point of the scale. These two items received mean rankings by the instructors in the previous study of 5.46 and 5.26 respectively. Similarly, the students gave relatively less importance than the instructors to the degree of isolation from fellow students and off-air interaction. Both these factors received mean rankings of 5.36 from the instructors and 4.56 and 4.88 respectively from the students.
Seven items received mean rankings of 3.0 or less. Five of these items were demographic in nature: size of community, age, employment level in the student's community, and, most strikingly, the gender of instructors and students, which received the lowest mean rankings of 1.80 and 1.87 respectively.
Table II shows the perceptions of two sub-groups or clusters of the distance students based on the scores of the Q-Sort items. Briefly, the Ward's (1963) method of cluster analysis partitioned two groups, and oneway analysis of variance provided the list of significant individual variables behind the group differences. That is, variables listed in each cluster scored significant higher means, for example, students' personality received a higher mean ranking by the 29 students comprising Cluster I.
Table III contains the mean rankings for each cluster on those variables that yielded significant F- values.
Students belonging to the first cluster can be typified as being student-centred; they gave significantly higher rankings to factors like student personality, student family obligation, student personal support, and so forth. Students in the second cluster, however, gave higher mean rankings to items that, in the main, directly relate to the instructional act, for example, instructor's knowledge of the structure of curriculum, instructor's aptitude for teaching, and quality of sound. This cluster can be typified as instruction-centred.
The perceptions of the students, as determined by the Q-Sort, compared with the perceptions of the instructors as measured by the administration of the same Q-Sort in the previous study. and shows significant Multivariate, Wilks Lambda = 0.18, overall F = 5.18, P .0001, and Univariate differences (significant F's) between instructor and student data sets. Note, multivariate statistics were obtained by recoding the missing values to 5, which was the mean of the 9 point scale; whereas the univariate or individual F's were computed by excluding the missing values (Tabachnick & Fidell, 1989, pp. 60–66). Thus, the students rated the first eight items in Table IV significantly higher than the instructors. These items relate to student characteristics and background, for example, student's attitude, student's personality, student's family support, and so on, and, as might be expected, they appear to be significantly more important to the students than to the instructors. Conversely, instructors generally ranked items related to instructor characteristics and background as being more important than did the students. An examination of items 9 to 15 in Table IV shows that the instructors gave significantly higher ratings than the students to instructor characteristics, with the exception of the instructor's knowledge of evaluation procedures and structure of the curriculum, which the students ranked more highly.
The instructors, predictably, tended to rank the more abstruse items, such as goals of education, as being more important than did the students. Out of the six items concerning goals of education, items 16 to 21, instructors rated five of them higher than the students. The single item goals of education, rated by the students higher compared to instructors, was personal growth as one of the goals of education.
The instructors appeared to give more importance to elements that impinge directly on the remote sites, items 22 to 27 in Table IV, than do the students. Although none of these six items was rated as being of relatively great importance to either instructors or students, in each case the instructors gave significantly higher rankings to the item than did the students. Surprisingly, the instructors ranked the degree of isolation and the group size per site as being more important factors than did the students, and also the frequency of visits by instructors was relatively more important to the instructors than to the students. On the other hand, four items that pertain to the nature and mores of the Institution, items 28 to 31, were ranked significantly higher by the students than by the instructors although, once again, none of the four items was ranked as being particularly important. Thus, the students gave higher mean rankings than instructors to whether the institution was single mode or dual mode, to the value of distance education in merit and promotion decisions, and to the attitude of the faculty or department to distance education.
Of the remaining nine elements for which significant differences were obtained between the mean rankings of instructors and students, two were technical and the rest administrative in nature. The instructors ranked the two technical factors - items 32 and 33 in Table IV as being relatively more important than did the students. Thus, both the ease of use of equipment students and the technical expertise of the bridge operator were concerning given higher mean rankings by the instructors than by the students. The students, on the other hand, tended to rate administrative factors that related to the nature of the course and its availability and applicability as being of more importance than did the instructors, items 34 to 37 in Table IV. Thus, the students ranked the scheduling of teleconference classes and whether the course is required or optional or non-credit as being of more importance than did the instructors. The students also gave more importance to the provision of library support services; however, the instructors attached more importance to the overall size of the class and the physical design of the teleconference space.
Table V summarizes the responses to the short questionnaire that was distributed to both the students in this study and the instructors who took part in the initial study.
The items contained in the questionnaires were developed from data collected during the initial study with instructors (Kirby & Chugh, 1992). The respondents were asked to indicate their agreement with each item via a five-point scale, (1 = strongly disagree to 5 = strongly agree). On the whole, neither the students nor the instructors indicated strong agreement or disagreement with any of the statements - with the exception of items 4 and 10 in Table V. Both groups disagreed with the statement that a qualification obtained via audio-teleconferencing instruction is of a lesser standard than one obtained through face-to-face instruction, and both also disagreed with the idea that the technology associated with audio-teleconferencing is a barrier to a true educational transaction. Significant differences between instructors and students were obtained for only two of the items. The instructors believed more strongly than the students that audio-teleconferencing instruction is as effective as face-to-face teaching and/or learning in a traditional classroom, and they disagreed more strongly than the students with the idea that audio-teleconferencing is a barrier to a true educational transaction. The data from the instructor responses to the questionnaires were further analyzed in light of the two types of instructors identified in the first study (Kirby and Chugh, 1992). Significant differences were found for two of the 10 items. Instructors who were typified as being more sympathetic with this form of mediated activity disagreed more strongly with the idea that qualifications obtained via audio-teleconferencing were of a lesser standard than instructors who were typified as being embedded in a traditional paradigm. (Mean rankings 1.70, 2.42, F=3.88, p=0.05.) The first type of instructor also disagreed more strongly than the second type with the idea that audio-teleconferencing was a barrier to a true educational transaction (mean rankings 1.37, 2.26, F=8.24, p=0.01).
The study reported in this paper examined the perceptions of the post-secondary students enrolled in audio-teleconferencing credit programs of factors in their instructional environment. The study also compared those perceptions with the perceptions of a group of audio-teleconferencing instructors, which were the subject of a previous paper (Kirby & Chugh, 1992). The instruments used in this study were a Q-Sort developed for use in the study on instructor perceptions and a short questionnaire to elucidate some perceptions that may have been buried in the Q-Sort data.
The salient elements in the instructional environment as seen by the students are factors that can be construed as being directly related to the outcomes of the teaching enterprise. Thus, not surprisingly, the instructor's aptitude for teaching, verbal skills, knowledge of the structure of the curriculum, together with factors such as motivation, aptitude, and personal goals of the student were rated most highly by the students. On the other hand, elements that can be interpreted as being peripheral to instructional outcomes received the lowest ratings. In these respects, there was considerable congruency between the results of the students in this study and those of the instructors surveyed previously. There were, however, some interesting differences when direct comparisons were made between the two groups.
Both groups tended to attach more importance to their own sets of characteristics than to each other's. Thus, the students tended to rate their own characteristics as being more important than those of the instructors and vice versa. This is hardly surprising since it is natural that both groups would tend to view the educational process from their own perspective. The instructors, however, ranked the more abstruse elements, such as goals of education, as being more important than did the students. They also ranked factors that could be interpreted as being related to the quality of the instructional experience more highly. Factors such as arrangements at individual remote sites, class size, and elements related to technical quality were ranked relatively higher by the instructors. On the other hand, the students rated elements related to the availability and value of a course as being more important. These results tend to underscore the importance of access to courses as an important issue for the students.
Findings of the previous study suggested that, from the instructor's viewpoint, distance education, including audio-teleconferencing, is a truly democratic form of education that tends to mask certain aspects of the instructor-student relationship, in particular, those that might contribute to a halo effect in a traditional setting. The relative lack of importance attached by the students to such factors as gender and personality tends to support this view. Indeed, both instructors and the students showed mild support for this viewpoint in their responses to the short follow-up questionnaire. On this basis, distance education may have a potential as a bias-free medium. Of immediate significance would be its utility in instruction in areas where there has been a history of gender bias, such as engineering, the physical sciences, and so forth.
The responses to the questionnaire tend to support the notion that experience with this medium builds support for it. Both instructors and students disagreed with the statements that qualifications obtained through audio-teleconferencing were inferior and that the medium was a barrier to a true educational transaction. There is an important lesson here for administrators struggling to convince policy makers of the medium's viability. It speaks to involving key individuals first-hand in the distance education operation.
The analysis of the instructors in the first study revealed two types of instructors. The first was more sensitized to the technology of the medium, and the other apparently corresponded to a more traditional paradigm of education. Parallels were drawn between these two clusters and an alleged paradigm shift discussed by Holmberg (1990), on the one hand, and Garrison and Shale (1990), on the other. The clusters identified in this study on the perceptions of students did not correspond to the typology of the instructors. It showed one group who attached comparatively more importance to elements related to the instructional process and another group who were more concerned with factors mainly related to the students' and to a lesser extent the instructors' background. Although no attempt was made in this study to relate this typology to other factors, it would be interesting to investigate its relationship to demographic and other variables. Lastly, it would be useful to determine important predicting variables that would distinguish between intragroup and intergroup differences, that is, between subgroups of students and instructors and between instructors and students, by means of discriminant analysis.
Bray, T. H., & Maxwell, S. E. (1985). Multivariate analysis of variance. Beverly Hills, CA: Sage.
Garrison, D. R., & Shale, D. G. (1990). Tilting at windmills? Destroying mythology in distance education. ICDE Bulletin, 24, 42–46.
Holmberg, B. (1990). A paradigm shift in distance education? Mythology in the making. ICDE Bulletin, 22, 51–55.
Kirby, D. M., & Boak, C. (1987). Developing a system for audio-teleconferencing analysis (SATA). Journal of Distance Education, 2(2), 31–42.
Kirby, D. M., & Boak, C. (1989). Investigating instructional approaches in audioteleconferencing classes. Journal of Distance Education, IV(l), 5–
19. Kirby, D. M., & Chugh, U. (1992). An investigation of instructors' perceptions of elements in audio-teleconferencing environment. Journal of Distance Education, VII(1), pp. 25–38.
Stephenson, W. (1953). The study of behaviour. Chicago: University of Chicago Press.
Tabachnick, B. G., & Fidell, F. S. (1989). Using multivariate statistics. New York: Harper & Row.
Ward, J. H. (1963). Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association, 236–244.
Wishart, D.(1987). Clustan users manual. St. Andrews: Computing Laboratory, University of St. Andrews.
David M. Kirby is a Professor and Dean of the Faculty of Continuing Education at the University of Calgary, Canada. His research interests include distance education and demographic and institutional research relevant to part-time university students.
Urmil Chugh holds an M.A. in Biological Science from the City College of The City University of New York, and M.E. Des. (Master in Environment Design) from the University of Calgary. As a researcher, she had undertaken a wide variety of research projects in the areas of social sciences, education and health sciences at the University of Calgary and with Provincial Government Departments.