The Relationship Between Comfort Speaking in Public, Gender, and Confidence

This research has two purposes: first, to look for a relationship between confidence and comfort speaking in public (in class, for instance), and second, to see if comfort speaking in public differs with gender.

Speaking in public is stressful to many people, and Paalhar (2004) cites that most people rate the fear of speaking in public as number one, higher than even the fear of death. One main reason people feel anxious about speaking in public is a lack of confidence, which often originates from the feeling that you are an inferior speaker compared to others, or that you know less about the topic in discussion (Paalhar, 2004).

Fassinger (1995) cites that speaking in class (a form of public speaking) is important, because it cultivates critical thinking in both discussion and lecture type classes. Weaver and Qi (2005) state that students, who participate actively in class, learn more than those who do not. They cite that participating actively promotes better memory retention (Weaver and Qi, 2005). Therefore, it would be worthwhile to ascertain what factors contribute to being comfortable speaking in class, to be able to focus on developing them and improving intelligence (Fassigner, 1995).

Previous studies on class participation and gender include Brady's and Eisler's (1999) study on gender in the college classroom, which analyzed faculty-student interactions and perceptions. The participants were students and instructors from 24 classrooms in 8 departments at a top-ranking university. The variables observed included genders of the student and instructor, class size, faculty belief of classroom equity, and the class' gender ratio. For exploratory analysis, the course's gender relevance was also included. Majority of the data analyses used multivariate analyses of variance (MANOVAs) and ANOVAs. The findings were that male and female students did not differ in classroom participation or perceptions, and instructors did not interact differently with them. This conflicted with their hypothesis that classrooms have gender-based stereotypes associated with them (Brady & Eisler, 1999).

Student perceptions, however, strongly correlated with the students' own behaviors and with instructor behaviors. Classroom interactions and student perceptions varied with demographic characteristics including instructor sex, class size, instructor monitoring of gender-race equity in the classroom, gender relevance of the course, and the sex ratio of the class. All of the sex differences were based mainly on student behaviors and were not necessarily influenced by behavior of instructors. This implies that students may have great influence on the kind of education they get (e.g., attention in class) that is comparable to the influence that instructors carry to call on students (Brady & Eisler, 1999).

This study is related to my research hypothesis. Brady and Eisler (1999) believe that gender-based stereotypes do not exist in universities and that student characteristics are what affect behavior. I am going to test whether a difference exists between genders and their comfort speaking in public, which could be a result of gender-based stereotypes. I am also going to see if confidence, a student characteristic, is related to their comfort speaking in public, which probably influences how students behave in class (e.g. how often they participate).

Another previous study, by Fassinger (1995), explored students' and professors' roles in classroom interaction. It noted that students do not get different treatment from professors based on gender, so the source of gender variance in participation could come from the students. It pointed out a clear trend in student behavior: males participate more in classes than females, who perceive themselves as less involved in class interaction. As cited by Fassinger (1995), a survey of 761 college students found that both genders have similar ranks of reasons for not speaking, but women are significantly more likely to say that they do not speak because they have poorly made ideas, lack knowledge about a subject, and fear looking dumb in front of peers. Males are significantly more likely to stay quiet, because they are unprepared or worry it will affect their grades. The purpose of this study was to explore the role of confidence, interest, and comprehension in students' classroom involvement.

Random professors who taught during the same time period were chosen from a private, mid-western liberal arts college. Fifty-one of the fifty-four professors agreed to have their classes surveyed during the 6th week of the semester. Students completed survey forms in class and were asked to respond regarding their present class. The surveys were given by eight trained students and took twenty minutes to complete. Two students chose not to do the ten-page survey. The questionnaire was pre-tested on sixty students. The classes had three to fifty students, with a mean of twenty. Most were female (60%); white; spoke English primarily (96%), and were nineteen years old. The total number of participants was 1,059.

The dependent variable was class participation, defined as any student comments offered or questions raised in class." The independent variables tapped three basic factors that could shape student participation: class traits, student traits, and professor traits. Class traits included size, gender distribution, emotional climate, interaction norms, participation grade, and interaction type. The student traits included confidence, preparation, comprehension, and interest. Confidence was measured through seven questions about students' fear of appearing dumb, being unable to organize thoughts quickly and express oneself clearly, feelings of intimidation and tenseness, and beliefs that one's view would offend others (alpha = 0.80). Professor traits were examined through students' perceptions of their professors' inclusiveness, approachability, feedback style, supportiveness, discussion style, and expertness.

The results agreed with the hypothesis, that student's characteristics may shape interaction. Males saw themselves as more confident and more involved in classroom interaction than did the females (p < .05). The females saw themselves as more prepared for class (p < .01), more interested in the subject matter (p < .05), and more interested in peers' comments and questions (p < .01). Confidence (a student trait) had the highest correlation with student participation (r = 0.52).

This study is related to mine, as its participants are similar (undergraduate college students). Its findings are a basis for my hypothesis, as they state that student characteristics affect class participation. I am going to see whether gender and confidence are related to a person's comfort speaking in public, which I assume is connected to class participation.

Another previous study is by Weaver and Qi (2005), and it is on college students' perceptions and class participation. Weaver and Qi (2005) got variables from their previous survey on teaching and learning at a medium, urban, Midwest public university. The final questionnaire had 233 items that assessed students' perceptions of issues related to their learning experiences. Since they relied on the survey and not direct observation, all variables are from self-reports, which may differ from observed reality. Participants were taken from the spring 2000 course catalog. Classes were purposively selected proportionately to the total number of courses from all departments of the six colleges, at all levels, and during the day or night. The total participants were 1,805 students, 16.6% of the population. Of the valid cases, three key demographic variables were chosen--age, sex, and ethnicity. The sample included 56% women and 44% men (compared to 54% and 46% in the population).

They made a path model to estimate how students' attributes directly and indirectly influence class participation. The dependent variable was class participation, measured by the self-reports of participation in classes: never, seldom, sometimes, usually, or always. This was thought to be more reliable than a measure that requires students to recall and specify the number of times they raise questions or respond within a time frame.

Their findings on the influence of gender on class participation were that gender has little or no effect on self-reported participation. This was inconsistent with their hypothesis, as they thought that males would report higher participation than women, since the classroom offered a "chilly" climate for women and thus favored "masculine" forms of communication: public arguments and displays of intelligence (Weaver & Qi, 2005). Gender had the smallest direct and total effect on participation (beta = 0.025 and 0.001, respectively) among all of their predictor variables.

Given Weaver and Qi's (2005) findings on gender and class participation, I have developed a hypothesis to test if a student's comfort speaking in public differs with gender, as I am curious to see if the results of my study will support or conflict with their findings.

Their other findings on confidence agreed with their hypothesis. Their model showed that confidence directly affected self-reported participation (beta = 0.167, p < 0.01). They interpreted this as confidence being affected by and sustaining the social structure of the classroom. I interpret this as confidence being important for class participation, a form of speaking in public. For this reason, I am going to test if confidence and comfort speaking in public are related.

Myaskovsky, Unikel, and Dew (2005) did a study on the effects of gender diversity on performance and interpersonal behavior in small work groups. One goal of this experiment was to observe the effects of gender diversity and solo status on group members' interpersonal behaviors. At the University of Pittsburgh, 288 students were recruited from Introductory Psychology classes and randomly assigned to a total of 96 three-person groups with different gender compositions (all men, all women, one woman and two men, or one man and two women). Most participants were White and ranged in age from 18 to 25. Groups were randomly assigned to either a male or female trainer.

Myaskovsky, et al. (2005) found differences in group members' interpersonal behaviors with respect to whether participants worked in mixed-gender or same-gender groups. Women were less talkative when solo than when belonging to the majority. In contrast, men were more talkative when solo than when they belonged to the majority. This was an unexpected finding. It could imply that different gender compositions in groups influence public speaking.

This study is related to my hypothesis. If differences between gender compositions in groups influence public speaking, then perhaps genders differ in how comfortable they are speaking in public.
My study will perform a correlation between confidence and comfort speaking in public (CSP). A self-rated 5-point scale ranging from "opposite of me" to "very much like me" will measure a person's confidence. A self-rated 5-point scale will measure CSP, ranging from "strongly disagree" to "strongly agree". A t-test will determine if a difference exists between genders and CSP. My first hypothesis is that confidence is positively correlated with comfort speaking in public (in class, for instance). My second hypothesis is that comfort speaking in public differs for each gender, with men being more comfortable than women.
Method

Participants

A total of 56 students at the University of Southern California (27 males and 29 females) were recruited, of which 54 were undergraduates and two were graduates. Each experimenter recruited two males and two females.
Materials

The Psyc 314 Tuesday morning laboratory class created a 20-item survey with questions on extraversion and school involvement (see Appendix A). The survey had seven open-ended questions, which asked for the number of hours the person spent on campus doing different activities, how many electives the person had taken at USC, and basic information such as the person's gender and age. It also included eight questions, where students were asked to rate how strongly they agreed or disagreed with a statement on a 5-point scale, such as how comfortable the person was with speaking in public. It had two questions, where students rated how well the words "outgoing" and "confident" described them, ranging from "the opposite of me" to "Just like me" (also a 5-point scale).
Procedure

Two female and two male USC students were approached by each experimenter and asked to fill out a survey for a Psyc 314 class. If they consented, then the survey was given. The researcher sat nearby as the participants filled out the survey. The participants were given as much time as they needed to complete the survey. They turned in the survey when completed, and the researcher thanked them for participating.
Results

A Spearman's rho rank order correlation was conducted on the scores of confidence and comfort speaking in public, r(56)=.359 p=.007 (two-tailed). Scores on confidence were positively and significantly correlated with comfort speaking in public.

A t-test was used to see if the genders differed on ratings of comfort speaking in public (CSP). The difference between CSP for males (M=3.741, SD=1.023) and females (M=3.103, SD=1.234) was significant, t(56)=2.095, p=.041. Males scored significantly higher on the CSP scale than females.
Discussion

A person's comfort in public speaking increases as self-confidence increases. This finding supports the Fassigner (1995) and Weaver and Qi (2005) results that confidence affects class participation.
Males are more comfortable speaking in class compared to females. This finding supports Fassinger's study (1995) and the idea she cited of college as a "cold climate" for females, but it conflicts with Weaver and Qi's (2005), whose results showed that gender has little impact on self-reported participation. It also conflicts with Brady and Eisler (1999), whose findings were that male and female students did not differ in classroom participation or perceptions.
One reason for this discrepancy could be that the sample used in my study was not as large and diverse as the ones used in Weaver and Qi's and Brady and Eisler's studies. Another difference could be attributed to the fact that Weaver and Qi's participants were from a public university, while mine were from a private one. Perhaps a public university is more liberal, which could explain the lower existence of gender stereotypes. A future study could include a survey with a question on political orientation.

The findings from this study also support the Myaskovsky, et al. (2005) finding that gender differences exist in group members' interpersonal behaviors with respect to whether participants worked in mixed-gender or same-gender groups. My findings would interpret the Myaskovsky, et al. (2005) results as connected to the gender difference in comfort speaking in public. Women could be less talkative when solo than when belonging to the majority, because they are more comfortable belonging to the majority. In contrast, men could be more talkative when solo than when they belong to the majority, because they are more comfortable when there is less competition around. These statements could be used as hypotheses for future studies.
This research only includes self-reports of comfort speaking in public. These reports could be inconsistent with actual participation in class, which is one measurable form of public speaking (e.g. number of times a person recites). Future studies could include observed reports of recitation by an outsider (i.e. not the teacher or students) on top of a more detailed survey of students' perceptions on public speaking. Participants in future studies should be gathered from the same classes, so that teachers and subjects can be controlled.

References
Fassinger, P.A. (1995). Understanding classroom interaction: students' and professors'

contributions to students' silence. Journal of Higher Education, 66, 82.
Weaver, R., & Qi, Z. (2005). Classroom organization and participation: college student's
perceptions. Journal of Higher Education, 76, 570-601.
Brady, K.L., & Eisler, R.M., (1999). Sex and gender in the college classroom: A
quantitative analysis of faculty-student interactions and perceptions. Journal of
Educational Psychology, 91, 127-145.
Myaskovsky, L., Unikel, E., & Dew, M.A. (2005). Effects of gender diversity on
performance and interpersonal behavior in small work groups. Sex Roles, 52,
645-58.
Paalhar, S.L., (2004). Gender differences in public speaking anxiety. Retrieved March 9,
2006, from http://clearinghouse.missouriwestern.edu/manuscripts/226.asp.