Puzzling Patterns: Exploring the Relationship between Confidence and Filler Words in Teenage Speech

Abstract

The present study investigates whether an adolescent’s confidence in their knowledge correlates to the frequency of filler words in their speech. High school participants (N=31) were presented with two mazes, Maze A to evoke more confidence and obstacle-ridden Maze B to evoke less confidence. Participants submitted Likert Scales regarding their confidence in completing each maze, then completed the mazes. They were then recorded while discussing each maze. Finally, their speech was transcribed and annotated for filler words, marking distinctions between filled pauses and discourse markers. Using this data, a Spearman Rank Correlation Coefficient test and a paired t-test were conducted to test the strength of correlation between filler word frequency and confidence. The results of the Spearman Rank test implied negligible correlation between the variables, r(29) = -.1, p = .184. The paired t-test, using an accepted significance level of 0.05, indicated no difference in the means of filler word frequency data before confidence is decreased (M=5.5, SD=2.8) and filler word frequency after confidence is decreased (M=5.7, SD=3.1), t(30) = .3, p = .740). Additional Spearman Rank tests and paired t-tests results on filled pause and discourse marker usage before and after variation in confidence implied that an adolescent’s confidence does not influence the type of the filler word they use. Due to sample size limitations, these conclusions cannot be generalized, so further research is needed, but this data lays the foundation for adaptations to school curriculums and societal perceptions of adolescents.