Relationships of linguistic and motivation variables with drinking outcomes following two mailed brief interventions

Susan E. Collins, Kate B. Carey, Joshua Smyth

Research output: Contribution to journalArticle

13 Scopus citations

Abstract

Objective: This study was a post hoc analysis of linguistic and motivation variables found in writing samples following the administration of two mailed brief interventions. Method: At-risk college drinkers (N = 100) received personalized normative feedback (PNF) or an alcohol education (AE) brochure via mail. Participants responded to open-ended questions describing their reactions to the information they received. The writing samples were then coded for linguistic characteristics using the Linguistic Inquiry and Word Count program and for proportions of self-motivational statements using a modified version of the Motivational Interviewing Skills Code. Results: Group comparisons indicated that the PNF group used a significantly higher percentage of first-person-singular and school-related words, whereas the AE group used a higher percentage of discrepancy, second-person and body-related words. Furthermore, the PNF group produced more language consistent with motivation to change than did the AE group. Hierarchical regressions testing mediation and moderation indicated that linguistic references to school and motivation moderated the group effect on changes in consumption during the heaviest drinking week. Further, although the group predicted reduction in heavy, episodic drinking, its effect was completely mediated by linguistic variables. Conclusions: Findings confirmed that PNF elicits distinct verbal responses that are associated with increased motivation and behavior change.

Original languageEnglish (US)
Pages (from-to)526-535
Number of pages10
JournalJournal of Studies on Alcohol
Volume66
Issue number4
DOIs
Publication statusPublished - Jul 2005

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Medicine (miscellaneous)
  • Psychology(all)

Cite this