Abstract
This study investigated the effect of a vicarious experience on the academic self-efficacy of graduate students enrolled in a statistics and research methods course. Participants (N = 39) completed a self-efficacy scale during the first two meetings of the course. Two weeks later, a portion of these students participated in a randomly assigned intervention to increase statistics self-efficacy. In the experimental condition, a former statistics student came to the class and explained her own math anxieties and outlined the behaviors that led to her personal success in the same course. Comparison students wrote about the characteristics of a successful student in the course without the experience of a peer model presentation. Analysis of pre- and postintervention academic self-efficacy indicated students in the peer model group showed a significant increase compared to the writing group. We discuss the potential of using live vicarious experience presentations to increase self-efficacy in psychology statistics courses.
Psychology instructors consistently face student anxieties, fears, and cognitive obstacles associated with taking statistics courses. Low self-efficacy with regard to math and statistics often affects performance (Bong & Skaalvik, 2003), leading to many studies exploring the links between self-efficacy and success in mathematics (Finney & Schraw, 2003; Friedel, Cortina, Turner, & Midgley, 2010; Harlow, Burkholder, & Morrow, 2006; Kennett, Young, & Catanzaro, 2009). Undergraduate and graduate psychology courses in statistics and research methods require innovative approaches to help students strengthen their own personal statistics self-efficacy during early stages of the course. Schunk (1991) defined academic self-efficacy as students’ beliefs in their own abilities to successfully perform a set of given academic tasks, such as passing a course. A multitude of studies identify self-efficacy as the strongest predictor of academic performance level (Bandura, Barbaranelli, Caprara, & Pastorelli, 1996; Ferla, Valcke, & Cai, 2009; Hackett, Betz, Casas, & Rocha-Singh, 1992; Lane & Lane, 2001; Lane, Lane, & Kyprianou, 2004; Lent, Brown, & Gore, 1997; Lopez & Lent, 1992; Multon, Brown, & Lent, 1991). Self-efficacy in academic contexts also predicts grade goals, cognitive strategy use, self-regulation, and intrinsic interest (Bong & Skaalvik, 2003).
Research suggests that the modeling of success through vicarious experience (Bandura, 2008; Lane et al., 2004) may have the most instant and direct influence on self-efficacy (Gist & Mitchell, 1992; Luzzo, Hasper, Albert, Bibby, & Martinelli, 1999). During a vicarious experience, a model informs and motivates by providing information about specific behaviors and strategies that led to personal success in that particular situation (Schunk, 2003). Witnessing the successful performance by a model supports the belief in one’s own ability to perform the task, such as passing a statistics course, successfully. Prieto and Meyers (1999) included vicarious learning in their assessment of teaching assistant training and supervision, but the variable was not specifically isolated as a predictor of self-efficacy. Harlow et al. (2006) investigated the impact of several pedagogical strategies, including semester-long peer mentoring, on self-efficacy within a quantitative psychology course. Although students’ reported statistics self-efficacy increased by the end of the term, instructors need efficient ways to increase academic self-efficacy during very early stages of these courses. Luzzo et al. (1999) examined vicarious experience by showing undergraduate students a 15 min video presentation of two graduate students describing their personal experiences with math and science. The video alone failed to significantly influence math and science self-efficacy among student viewers and the researchers recommended live models for intervention success. Although this particular approach utilized vicarious learning, the models represented a higher level of education than the viewers, which may have weakened the effectiveness of the intervention. In general, greater similarity between the observer and the model contributes to greater persuasiveness of the model (Bandura, 1997, 2008; Schunk, 1987).
In their influential paper distinguishing academic self-concept from academic self-efficacy, Bong and Skaalvik (2003) called for additional research on instructional approaches that will increase student self-efficacy in the context of specific courses. Psychology instructors may benefit from the use of peer models to enhance self-efficacy among students, especially within courses that produce greater levels of anxiety among students. The current study measured the impact near the beginning of the semester of a live vicarious experience intervention, a presentation by a former student that successfully completed the course, on students’ perceived academic self-efficacy with regard to successful completion of a statistics course. We expected students witnessing a live presentation from a similar peer model to report a significant increase in academic self-efficacy compared to those with no peer model. The findings will aid teachers of psychology aiming to increase student perceptions of self-efficacy with respect to academic performance in a particular course.
Method
Participants
Participants included 64 students from three sections of a master’s-level research design and statistics course at a medium-sized state university. The sample consisted of 54 females and 10 males. The groups’ average age was 28 years (range: 21–63). Students identified as White (61%), African American (6%), Hispanic/Latino (19%), or other (14%). Students did not receive any extra credit or other incentive for their participation.
Instruments
The self-efficacy scale (Cronbach’s α = .94) used in this study was developed based on a measure from Lane and Lane (2001). Our scale measured the confidence on competencies needed to achieve success in the research design and statistics course. We asked participants to answer eight self-efficacy questions on a 10-point Likert-type scale ranging from (1) no confidence to (10) total confidence. Sample items from the scale included: “You have sufficient math ability to meet the demands of the course” and “You can manage your time to meet the demands of the course.”
Procedure
At the beginning of the semester, we asked all students in the class to complete informed consent forms along with the self-efficacy scale. Participants were not told about any future measurements. Two weeks after this pretest, we randomly assigned students to experimental and control group interventions and placed these groups in separate rooms. A previous research and statistics course graduate student made a 10 min presentation to the experimental group. The female peer model presenter, close to the average participant age, described her performance level as similar to the majority of students enrolled in the course. The presenter discussed her experiences in the course, how she managed her time, her study methods, and how she managed stress related to the course. She also outlined her eventual success in the course. To make the presentation more realistic, the student did not have a script but rather an outline of what she would say. Questions or comments were not allowed after her vicarious presentation. After the presentation, students completed a posttest self-efficacy scale identical to the pretest. It is possible that some participants knew or knew of the student, but we believe most did not.
In a separate room, control group participants were told to imagine a successful student in a research and statistics course. While reflecting on this successful peer, control group participants wrote for up to 10 min about this student’s time management, study habits and methods, and stress management skills that lead to success in the course. After receiving the instructions, students wrote about the successful student in whatever form they wanted. Following the reflective writing exercise, students completed the posttest self-efficacy scale.
Results
We conducted a 2 × 2 factorial ANOVA to determine whether testing time (pre-intervention vs. postintervention) and group (control vs. experimental) had an effect on participants’ overall self-efficacy ratings. We used the pre-intervention ratings to control for initial levels of self-efficacy as well and to determine any changes in self-efficacy. As expected, analyses indicated no significant main effect of testing time, F(1, 37) = 0.07, ns, or group type, F(1, 37) = 0.11, ns. In addition, the ANOVA revealed a significant Testing Time × Group interaction, F(1, 37) = 9.38, p = .004, partial η2 = .20. A marginally significant increase in academic self-efficacy ratings occurred for students in the experimental group from pre- (M = 6.66) to postintervention (M = 7.07), t(19) = –2.01, p = .06, d = 0.45. 1 Control group participants exhibited a significant decrease in self-efficacy from pre- (M = 6.96) to postintervention (M = 6.48), t(18) = 2.32, p = .03, d = 0.53.
Discussion
The findings suggest that students may improve self-efficacy by hearing about previous course experiences from former successful students in person. In addition, a presentation from an average student, as opposed to the “star pupil,” may be especially useful to increase the effectiveness because current students recognize similarities between themselves and the model. In fact, to counter low self-efficacy in students, it may be important for the model to discuss experiencing low academic self-efficacy upon first entering the course. In the control condition, student reflections and writing about an imagined successful student failed to improve self-efficacy and even indicated a possible decreased self-efficacy among participants. When imagining a successful student, participants may think of somebody quite different from themselves and consequently assume that the fictitious model possessed unique qualities necessary for success. The decreased self-efficacy among control students may also be influenced by learning more about the course requirements and expectations in the first two weeks of the course, as well as initial encounters with the textbook and statistics assignments.
In this study, providing a vicarious experience from a successful model, who students perceive as similar to themselves, led to an increase in self-efficacy. Although this study focused on master’s-level students in a research and statistics course, the live vicarious experience peer model presentation may also enhance academic self-efficacy in both undergraduate and graduate psychology statistics and research methods courses. However, we also have tentative evidence that a poor presentation by a vicarious model may reduce academic self-efficacy. Furthermore, the medium itself (video vs. live) may influence student levels of academic self-efficacy. The live interaction could provide a more realistic or authentic presentation, leading to greater persuasion. Viewing a video (Luzzo et al., 1999) may feel more distant and more difficult to connect with personally. In contrast, a live presentation provides students with more opportunities to make connections and find similarities between themselves and the former student. Future research might explore the video versus live presentation methods and measure perceptions of persuasiveness. Additionally, further research could examine the role that perceived similarity to the peer model plays in improving self-efficacy.
Several aspects of the study design and sample selection limit broad application of the findings. Given that students completed the pre- and posttest surveys in class, self-report bias may have influenced the responses. Although none of the researchers collecting the data taught the courses involved in data collection, results may still be affected by demand characteristics; however, given that both groups had an intervention, it is unclear why the students in one intervention would be affected more by demand characteristics than another. Furthermore, it is unclear whether this finding is limited to anxiety-producing courses such as statistics or if this finding can be broadened to other areas of academic self-efficacy in other courses or even degree attainment.
Footnotes
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The authors received no financial support for the research, authorship, and/or publication of this article.