The three objectives of this dissertation were to develop a question type model for predicting linguistic features of responses to interview questions, create a tool for linguistic analysis of documents, and use lexical bundle analysis to identify linguistic differences between fraudulent and non-fraudulent financial reports. First, The Moffitt Question Type Model (MQTM) was developed to aid in predicting linguistic features of responses to questions. It focuses on three context independent features of questions: tense (past vs. present vs. future), perspective (introspective vs. extrospective), and abstractness (concrete vs. conjectural). The MQTM was tested on responses to real-world pre-polygraph examination questions in which guilty (n = 27) and innocent (n = 20) interviewees were interviewed. The responses were grouped according to question type and the linguistic cues from each groups' transcripts were compared using independent samples t-tests with the following results: future tense questions elicited more future tense words than either past or present tense questions and present tense questions elicited more present tense words than past tense questions; introspective questions elicited more cognitive process words and affective words than extrospective questions; and conjectural questions elicited more auxiliary verbs, tentativeness words, and cognitive process words than concrete questions. Second, a tool for linguistic analysis of text documents, Structured Programming for Linguistic Cue Extraction (SPLICE), was developed to help researchers and software developers compute linguistic values for dictionary-based cues and cues that require natural language processing techniques. SPLICE implements a GUI interface for researchers and an API for developers. Finally, an analysis of 560 lexical bundles detected linguistic differences between 101 fraudulent and 101 non-fraudulent 10-K filings. Phrases such as "the fair value of," and "goodwill and other intangible assets" were used at a much higher rate in fraudulent 10-Ks. A principal component analysis reduced the number of variables to 88 orthogonal components which were used in a discriminant analysis that classified the documents with 71% accuracy. Findings in this dissertation suggest the MQTM could be used to predict features of interviewee responses in most contexts and that lexical bundle analysis is a viable tool for discriminating between fraudulent and non-fraudulent text.

As increasing number of organizations are using virtual teams, communication scholars have started to pay more attention to these relatively new forms of work. Past studies explored interpersonal (i.e., trust, attraction) and group dynamics (i.e., conformity, subgrouping) in virtual teams. Despite the documented effects of interpersonal behaviors and social categories on virtual group dynamics, there is a substantial gap in how these two factors influence language use in virtual teams. To shed light on this neglected area of research, this dissertation examined how teammates’ interpersonal behaviors and social categories affected language use in virtual team collaborations. 164 participants interacted in four-person teams using a synchronous chat program. The age of participants ranged from 18 to 24. 58% of participants were female and 42% were male. Participants used Windows Live Messenger to complete Straus & McGrath’s (1994) decision making task. Upon completing the task, participants filled out social attraction and social identification scales to be used for manipulation checks. Decision making sessions for each group were saved and Linguistic Inquiry and Word Count Program (LIWC) was used to examine language use. Linguistic style accommodation was measured using language style matching (LSM) metric. LSM measured the degree to which group members used similar language patterns. It was calculated by averaging the absolute difference scores for nine function word categories generated by LIWC. Similarly, linguistic markers such as word counts, negations, assents, and pronouns were acquired through LIWC output. The results suggested that having a dissenting member in the group was associated with higher linguistic style accommodation compared to having an assenting member. This result contradicted with the assumptions of communication accommodation theory (Giles, Mulac, Bradac, & Johnson, 1987), yet provided evidence for the validity of minority influence theory (Moscovici, Lage, & Naffrechoux, 1969) in virtual teams. Unexpectedly, there was no significant effect of social categories on linguistic style accommodation. The results also showed that negative behaviors were strongly associated with increased word counts, negations and the second person singular pronouns, whereas positive behaviors were associated with increased use of assents, tentative language, first person plural and singular pronouns. / text