Facebook and impressions of new roommates in the transition to college: The impact of discrepancies between online and offline roommate impressions on the development of roommate relationships among first year students

This dissertation is an attempt to explore first year college students' Facebook use in association with their relationship development with their previously unacquainted roommates. Survey data indicated that it is very common for freshmen to look up their roommates on Facebook after receiving their roommate assignment from school. Being able to get an idea of who the roommate is helps reduce a student's uncertainty about living with a complete stranger. Both the uncertainty reduction theory (URT) and the predicted outcome value (POV) theory provide a solid theoretical framework to predict students' information-seeking behaviors on Facebook. Social information processing (SIP) theory and hyperpersonal model are proved to be helpful in explaining students' impression formation process on Facebook. This research predicts that information-seeking behaviors as well as the impressions formed based on information available on Facebook will reduce students' uncertainty about the roommates. Moreover, the study aims to take a further step by applying expectancy violations theory to investigate whether the discrepant impressions formed between Facebook and offline experience have an impact on students' level of uncertainty upon move-in with the roommates, and more importantly, the impact on the development of roommate relationship closeness.;A combination of a three-wave survey method and in-depth interviews with 19 students was used for this study. Survey data were collected at three different time points: (1) two weeks before college freshmen moved in with their roommates; (2) one week after they moved into the dormitory; and (3) seven weeks after living together with the roommates. The interviews were conducted after the three-wave survey was completed.;Statistical analyses using multiple linear regressions, multiple analysis of variance, and mixed-design ANOVA were applied for the hypotheses testing. The findings were mostly consistent with the hypotheses: (A) before moving in with the roommates, incoming first year students' uncertainty level was affected by how often they interacted with the roommates on Facebook, how many channels they used to communicate with the roommates, and their impressions of roommates' appearance and task attractiveness; (B) freshmen who formed positive initial Facebook impressions engaged in more information-seeking behaviors and had greater certainty than those who formed negative initial impressions of the roommates; (C) upon moving in with the roommates, students' uncertainty level was influenced by their offline impressions of the roommates' social and appearance attractiveness; (D) discrepancies between initial Facebook and offline impressions produced significant group differences in students' level of uncertainty and relational outcomes with the roommates; and (E) students' uncertainty and the impressions of roommates change over time and among groups.