Role of anxiety on graduate cooperative groups.

Abstract

This study investigated the extent to which cooperative group members' anxiety levels predict the quality of an article critique project, and the degree that heterogeneity of anxiety levels is related to this group outcome. Participants were 109 graduate students enrolled in a research methodology course. Groups (n = 32) formed the unit of analysis. Findings revealed that groups attaining the lowest scores on an article critique assignment tended to report the highest anxiety levels and to be the most heterogeneous with respect to research anxiety.

**********

Johnson, Johnson, and Smith (1991a) define cooperative learning as "the instructional use of small groups so that students work together to maximize their own and each other's learning" (p. III). The initial theory underlying cooperative learning posited by Slavin (1994) comprised two elements, positive interdependence and individual accountability. Johnson and Johnson, and their colleagues (Johnson et al., 1991a; Johnson, Johnson, & Smith, 1991b), recommended incorporating in cooperative learning groups the following five-component model to maximize performance outcomes: (a) positive interdependence, (b) face-to-face promotive interaction, (c) individual accountability, (d) social skills, and (e) group processing.

Cooperative learning implemented at the primary and secondary levels of education for several subject areas has been found to promote higher performance levels in contrast to other instructional techniques (Johnson, Maruyama, Johnson, Nelson, & Skon, 1981; Slavin, 1994). Moreover, cooperative learning affects positively student attitudinal outcomes, such as elevating students' levels of self-esteem, self-efficacy, productivity, and motivation in relation to school and learning (Crooks, 1988; Johnson, Johnson, & Maryuma, 1983).

Although cooperative learning techniques are utilized at the college level in graduate- level courses, evaluative studies have not been conducted at this level to the same degree that evaluation has occurred at the primary and secondary levels (Slavin, 1991). Further, there is limited empirical research evaluating the impact of these techniques on graduate students' instruction and learning outcomes (Onwuegbuzie, Collins, & Elbedour, 2003) and the degree that group characteristics affect students' performance in cooperative groups (Onwuegbuzie & Collins, 2002). Recent research evaluating the impact of cooperative learning on performance outcomes indicates that graduate student participation in cooperative learning activities results in an increase in the number of meaningful learning opportunities (Collins & Onwuegbuzie, 2001; Onwuegbuzie, 2001; Onwuegbuzie et al., 2003; Onwuegbuzie & DaRos-Voseles, 2001; Wilson, 1998). DaRos-Voseles, Onwuegbuzie, and Collins (2003) found that graduate students' levels of perfectionism play a role in determining cooperative group outcomes. However, more empirically-based studies are needed that examine the impact of group characteristics on performance outcomes (Onwuegbuzie & Collins. 2002).

One feature of cooperative learning groups at the graduate level that was identified in Onwuegbuzie and DaRos-Voseles' (2001) qualitative inquiry was that some groups display higher levels of anxiety than do other groups. However, Onwuegbuzie and DaRos-Voscles were not able to investigate whether levels of anxiety experienced in each group predicted group outcomes. This is the goal of the present research. Specifically, the present investigation examined the role of research anxiety in predicting performance of cooperative learning groups in graduate-level research methodology courses. Research anxiety has been defined as an apprehensive state that impedes individuals' understanding and application of research-related activities associated with scientific methods, thereby impacting their confidence to develop and implement credible research (Kracker, 2002; Kracker & Wang, 2002; Onwuegbuzie, 1997a, 1997b, 1997c).

Students' levels of research-related anxiety have been found to impact negatively undergraduate students' perceptions about research (Kracker, 2002; Kracker & Wang, 2002) and graduate students' attitudes about conducting research related activities in the context of utilizing library and Internet resources (Collins & Veal, in press) and developing research proposals (Onwuegbuzie, 1997a, 1997b, 1997c). Onwuegbuzie and DaRos-Voseles (2001) reported that placing graduate students in cooperative learning groups lessened some groups' research-related anxiety levels. In stark contrast, qualitative analyses indicated that research-related anxiety levels of some groups were elevated by working cooperatively in groups. Thus, future research in this area can help to resolve this apparent contradiction.

The present study's focus on the role of anxiety as a characteristic impacting cooperative group outcomes is based on Wine's (1980) Cognitive-Attentional-Interference theory of test anxiety. According to this theory, in the context of processing information, an individual's level of anxiety impels cognitive interference that is characterized as a shift in a person's focus from task-relevant to task-irrelevant thoughts and actions leading to a diminished performance outcome. This link between anxiety and cognitive interference in the context of cooperative groups is supported by Onwuegbuzie and DaRos-Voseles (2001), who reported that cooperative groups that demonstrated more self-regulatory behaviors and heightened metacognitive awareness of the process of working cooperatively affected positively overall group functioning. Thus, it was hypothesized that anxiety would have a debilitative effect on cooperative group outcomes.

Method

Participants Participants were 109 graduate students from a number of disciplines (e.g., early childhood, elementary education, middle grades, secondary education, and psychology) who were enrolled in five sections of an introductory-level research methodology course at a midsouthern university. These students formed 32 groups ranging in size from 2 to 5 (M = 3.58, SD = 0.99). The same instructor taught all sections of the research methodology course, thereby minimizing any implementation threat to internal validity resulting from differential selection of instructors (Onwuegbuzie, 2003).

Setting All graduate students enrolled in educational degree programs were required to take the introductory-level research methodology course. The semester-long (i.e., 16-week) research methodology course involved classes that took place once per week for three hours. The fact that all classes were held at the same time in the evening (i.e., 5 pm to 8 pm) minimized any implementation threat to internal validity resulting from differential time of day (Onwuegbuzie, 2003).

Article Critique A major course requirement that was undertaken via cooperative learning groups involved a detailed written critical evaluation of a published research report (i.e., article critique). The primary goal of the article critique was to provide an opportunity for students to develop skills in evaluating published research articles utilizing principles of the scientific method.

Formation of Cooperative Learning Groups On the first day of class, students, in turn, were asked to introduce themselves to the class, delineating their major, educational attainments and aspirations, current professional status, and interests. Following these introductions, students were asked to form groups comprising 3-6 students. Group formation was guided by asking students to choose group members based on similar majors, professional background, and proximity to each other's homes. These criteria for group assignment were not directly related to aptitude or ability. Such assignment of groups by preferences is referred to as a modified stratified random assignment (Johnson & Johnson, 2000).

Base Groups The cooperative learning group that was utilized involved the use of base groups (Smith, Johnson, & Johnson, 1992). The aim of these base groups was to promote stable membership whose foremost responsibility was to provide each member of the group the support, encouragement, and assistance as needed to comprehend course content. In addition, the cohesiveness provided by membership in the group was (a) to promote the successful completion of the course assignments and (b) to prepare students for the in-class individual examinations. Students were encouraged to stay together during the entire course. Additionally, students were expected to take notes for and to distribute any instructor handouts to any group member who was unable to attend a class session. Each base group undertook one research proposal and one article critique.

Instruments To measure students' levels of research anxiety, the Research Anxiety Rating Scale (RARS) was selected for the present study. The RARS (Onwuegbuzie, 1996) is a 45-item, 5-point Likert-format instrument designed to measure levels of anxiety experienced by students enrolled in research methodology courses. Scores for the total scale ranged from 45 to 225, with a high score indicating a high level of anxiety. For the present investigation, Cronbach's coefficient alpha was .96 for scores pertaining to the RARS instrument. For the article critique, three rubrics were used. The first rubric consists of a 5-point Likert-format scale (1 = strongly disagree, 2 disagree, 3 = neutral, 4 = agree, 5 = strongly agree) that was designed to provide a score for the summary of the selected research article. This scale contains 35 items (e.g., "The conceptual/theoretical framework is summarized adequately"), such that scores range from 35 to 175. The second rubric, also consisting of 5-point Likert-format scale (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree), assesses how accurately the 150-item reviewer checklist described above is completed. Each response on the reviewer checklist is rated on the 5-point Likert-format scale, such that the second rubric contains 150 items, whose scores range from 150 to 750. The third rubric, also a 5-point Likert-format scale (1 = strongly disagree, 2 = disagree, 3 neutral, 4 = agree, 5 = strongly agree), was designed to assess the narrative for the critique section of the article. This rubric contains 50 items that evaluate all components of the critique section (i.e., title, abstract, introduction/literature review, methodology, results, discussion), such that scores range from 50 to 300. This third rubric also assesses the extent to which the critique section is free from grammatical and typographical errors and follows APA guidelines. Scores from the three rubrics were converted into percentages. From these percentages, a final score was derived using the following weighting scheme: 35% for the summary rubric, 25% for the reviewer checklist, and 40% for the critique narrative. Thus, each article critique received a group score on a 100-point scale. Because of the length of the rubrics, each article critique typically took between 6 hours and 7 hours to score. The 32 article critiques took approximately 16 days to grade. As a result, groups were required to turn in these assignments three weeks before the end of the semester (i.e., on Week 13 of the semester).

Analysis For each group, the mean and standard deviations pertaining to students' scores on the RARS were computed. This generated 32 group scores. That is, group was used as the unit of analysis. Utilizing groups themselves as the unit of analysis, rather than the individual scores, decreases the possibility of the statistical independence assumption being violated and systematic error being created (McMillan, 1999).

Results and Discussion

Using group as the unit of analysis indicated a statistically significant relationship between scores on the article critique and (a) within-group mean RARS anxiety scores (r = -.35, p < .05), and (b) within-group standard deviation of the anxiety scores (r = -.36, p <.05). Using Cohen's (1988) criteria, these associations were moderate.

The first purpose of this study was to investigate the extent to which cooperative group members' levels of anxiety predict group learning outcomes, as measured by the quality of an article critique project. Findings revealed that cooperative groups that attained the lowest scores on an article critique assignment, on average, tended to report the highest levels of research anxiety. This result is consistent with Wine's (1980) Cognitive-Attentional-Interference theory of test anxiety. Specifically, in accordance to this theory, an individual's anxiety level creates cognitive interference, leading to diminished performance outcomes. This is consistent with the finding of Onwuegbuzie, Slate, Patterson, Watson, and Schwartz (2000) that anxiety is an important predictor of performance in research methodology courses. Thus, it is likely that the cumulative effect of the groups' levels of anxiety debilitated their performance levels. Moreover, because anxiety debilitates individual performance in research methodology courses (Onwuegbuzie, 1997a), the present finding regarding the role of anxiety on group outcomes helps to explain the Matthew effect found by Onwuegbuzie et al. (2003), in which groups that contained higher-achieving students on an individual level tend to produce better group outcomes than do their lower individual-achieving counterparts. Simply put, groups containing students with the highest levels of anxiety are at risk of underachievement.

The second purpose of the study was to assess the degree that heterogeneity (i.e., variability of anxiety scores) is related to the quality of the group projects. Results indicated that cooperative groups that attained the lowest scores on an article critique assignment tended to be the most heterogeneous with respect to research anxiety. Thus, it appears that wide variability pertaining to group members' levels of anxiety may lead to conflicts that act as barriers toward working collaboratively within a group context. These results are consistent with DaRos-Voseles et al.'s (2003) interpretation that group cohesiveness is enhanced when members share similar perfectionistic standards about creating and evaluating a group-developed project. This finding suggests that levels of anxiety appear to play an important role among graduate students with respect to the performance of cooperative learning groups. Consequently, instructors who use cooperative learning techniques among graduate students should bear this in mind, and perhaps take steps to minimize levels of anxiety so that its impact would be reduced.

References

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New York: John Wiley.

Collins, K. M. T., & Onwuegbuzie, A. J. (2001, April). Group heterogeneity and performance in graduate-level educational research courses: The role of aptitude by treatment interactions and Matthew effects. Paper presented at the annual conference of the American Educational Research Association (AERA), Seattle, WA.

Collins, K. M. T., & Veal, R. E. (in press). Off-Campus adult learners' levels of library anxiety as a predictor of attitudes toward the Interact. Library & Information Science Research.

Crooks, T. (1988). The impact of classroom evaluation practices on students. Review of Educational Research, 58(4), 438-481.

DaRos-Voseles, D. A., Onwuegbuzie, A. J., & Collins, K. M. T. (2003). Graduate cooperative groups : Role of perfectionism. Academic Exchange Quarterly, 7(3), 307-311.

Johnson, D. W., & Johnson, F. P. (2000). Joining together: Group theory and group skills (7th ed.). Boston, MA: Allyn and Bacon.

Johnson, D. W., Johnson, R. T., & Maryuma, G. (1983). Interdependence and interpersonal attraction among heterogeneous and homogeneous individuals: A theoretical formulation and a meta-analysis of the research. Review of Educational Research, 53, 5-54.

Johnson, D. W., Johnson, R. T., & Smith, K. A. (1991a). Cooperative learning: Increasing college faculty instructional productivity. Washington, DC: The George Washington University. (ASHE-ERIC Higher Education Report No.4).

Johnson, D. W., Johnson, R. T., & Smith K. A. (1991b). Active learning: Cooperation in the college classroom. Edina, MN: Interaction Book Co.

Johnson, D. W., Maruyama, G., Johnson, R., Nelson, D., & Skon, L. (1981). Effects of cooperative learning, competitive, and individualistic goal structures on achievement: A meta-analysis. Psychological Bulletin, 89, 47-62. (University Microfilms No. 69-6443).

Kracker, J. (2002). Research anxiety and students' perceptions of research: An experiment. Part 1. Effect of teaching Kuhlthau's ISP model. Journal of the American Society for Information Science and Technology, 53(4), 282-294.

Kracker, J., & Wang, P. (2002). Research anxiety and students' perceptions of research: An experiment. Part 11. Content analysis of their writings on two experiments. Journal of the American Society for Information Science and Technology, 53(4), 295-307.

McMillan, J. H. (1999). Unit of analysis in field experiments: Some design considerations for educational researchers. (ERIC Document Reproduction Service No. ED 428135)

Onwuegbuzie, A. J. (1996). Development of the Research Anxiety Rating Scale. Unpublished manuscript, University of Central Arkansas.

Onwuegbuzie, A. J. (1997a). Writing a research proposal: The role of library anxiety, statistics anxiety, and composition anxiety. Library & Information Science Research, 19, 5-33.

Onwuegbuzie, A. J. (1997b). The teacher as researcher: The relationship between enrollment time and achievement in a research methodology course. Reflection and Research, 3(1) [On-line]. Available: http://www.gonzaga.edu/rr/v3n1/tony.html

Onwuegbuzie, A. J. (1997c). The teacher as researcher: The relationship between research anxiety and learning style in a research methodology course. College Student Journal, 31,496-506.

Onwuegbuzie, A. J. (2001). Criterion-related validity of checklists and rubrics in educational research courses. Unpublished manuscript, Valdosta State University, Valdosta, GA.

Onwuegbuzie, A. J. (2003). Expanding the framework of internal and external validity in quantitative research. Research in the Schools, 10, 71-90.

Onwuegbuzie, A. J., & Collins, K. M. T. (2002, April). Performance of cooperative learning groups in graduate-level educational research courses: The role of social interdependence. Paper presented at the annual meeting of the American Educational Research Association, New Orleans, LA.

Onwuegbuzie, A. J., Collins, K. M. T., & Elbedour, S. (2003). Aptitude by treatment interactions and Matthew effects in graduate-level cooperative learning groups. Journal of Educational Research, 96, 217-231.

Onwuegbuzie, A. J., & DaRos-Voseles, D. A. (2001). The role of cooperative learning in research methodology courses: A mixed-methods analysis. Research in the Schools, 8, 61-75.

Onwuegbuzie, A. J., Slate, J. R., Patterson, F. R. A., Watson, M. H., & Schwartz, R. A. (2000). Factors associated with achievement in educational research courses. Research in the Schools, 71(1), 53-65.

Slavin, R. E. (1991). Synthesis of research on cooperative learning. Educational Leadership, 48, 71-82.

Slavin, R. E. (1994). Cooperative learning: Theory, research, and practice (2nd ed.). Boston: New Jersey: Allyn & Bacon.

Smith, K., Johnson, D. W., & Johnson, R. T. (1992). Cooperative learning and positive change in higher education. In A. Goodsell, M. Maher, & V. Tinto (Eds.), Collaborative learning: A sourcebook for higher education (pp. 34-36). University Park, PA: National Center on Postsecondary Teaching, Learning and Assessment.

Wilson, V. A. (1998, November). A study of anxiety in graduate students in an introductory educational research course. Paper presented at the annual meeting of the Mid-South Educational Research Association, New Orleans, LA.

Wine, J. (1980). Cognitive-attcntional theory of test anxiety. In I.G. Sarason (Ed.), Test anxiety: Theory, research, and applications (pp. 349-385). Hillsdale, N J: Erlbaum.

Collins is an assistant professor. Onwuegbuzie is an associate professor in the Department of Educational Measurement and Research. DaRos-Voseles is an associate professor.