Context
Although there is an increased interest in education in virtual worlds, there is relatively little research that supports our understanding of the facets of interactions in virtual environments. Much of the recent and promising research has focused directly on avatar creation and reactions to varying avatars. For instance, Black et al. (2009) found there were varying conditions under which students would change the look of their avatars. Beck (2009) provided evidence that teachers predicted different success rates of avatars based on color and gender although avatars had similar test scores and backgrounds. Our own experience has provided anecdotal evidence that students seem to trust older avatars than they do younger counterparts. This area of study is critical for understanding teaching and learning in virtual worlds. The premise is that students might react differently given avatars chosen by instructors; teachers might view students differently based on avatars they have chosen.
Questions
The purpose of this study was to determine whether students differed in their opinions of an instructor given a simple change in represented age of the avatar. There were three related research questions:
1) Is there a difference between students' comfort level in interacting with an older or younger avatar?
2) Is there a difference between students' self-reported learning in interacting with an older or younger avatar?
3) Is there a difference between students' perceptions of instructor intelligence in interacting with an older or younger avatar?
Methods
Thirty-six students in two sections of an introductory course at a Northeastern university participated in the study. At the beginning of each of the two different classes, students were told about the study and given the informed consent. Informed consent was obtained and collected. Students who wished to participate were shown a video of how to build a simple object in the virtual world Second Life.
There was one video shown per class. Both videos had identical content and an identical script. One of the classes, however, was shown a video that had an avatar professor with a young, male appearance. The second class was shown a video that had an avatar professor with an older, male appearance. Each class watched their video and then completed a survey. The survey consisted of 12 questions regarding students' perceptions of the instructor and the training they had just received. Students were asked to strongly agree or disagree with the 12 statements. Analyses were run on the resulting scores.
Results. The mean scores from the study and the questions used are provided in Table 1. Due to formatting issues, Table 1 is not included here. Tables and illustrations will be provided during the presentation.
T-tests were also run on the differences between the groups. Four questions demonstrated a significant difference in the mean scores between the groups (P<.05): questions 9, 10, 11, and 12.
Discussion
Overall, students in both groups seemed to feel positive about the instructor (Q1, Q2, Q10, & Q11), comfortable with the instructor (Q3, Q4, & Q7), and felt like they had learned from the content provided (Q8 & Q9). Neither group seemed to feel comfortable about having a real life conversation with the instructor (Q5). However, in every answer, students in the group with the older, male avatar seemed to feel more positive about their experience. Students with the older, male avatar for an instructor felt significantly stronger about what they learned (Q9), the instructor's intelligence (Q10 & Q11), and perhaps most importantly, whether they would take a class from that instructor again.
Conclusion
Research on age and instructors has provided some evidence that students seem to value the experience of older instructors (Buzan, 1972). Thus, it is not surprising to replicate these results in virtual worlds. This is important for teaching and learning online because in real life, I cannot mask my age. However, in a virtual world, I can become older or younger with the click of a button. Future research needs to build upon this work to determine what happens in long term teaching scenarios rather than in simple video instructions.
Limitations
Students selected for this project were from a regional, rural campus, which could have impacted their response. The avatars used in this study were male; gender, race and age all might have some impact on learning, which needs to be studied. Finally, there were only two groups of 18; future research should build upon this study by adding a larger population.
