Analisis Statistika dalam Riset-Riset Psikologi Komunikasi: Sebuah Studi Literatur

Abstract

The increasing number of studies in psychology on communication demanded the use of more relevant statistical analysis techniques that can meet the needs for these studies. Conventional analysis techniques, such as ones that were based on linear model, assume that the observation are independent. Such assumption were often violated in studies on communication. Therefore, it was imperative to develop statistical analysis techniques that could address problems related to data with dependent observation either in communication events. Such efforts could not be separated from literature studies addressing several research questions: (1) What kind of research questions that frequently arose in communication in Psychology?, (2) What statistical analysis techniques that were commonly used to answer the questions in such studies?, (3) What potential statitical analysis techniques that could be used in the studies?, and (4) How were the discrepancies between the needs in communication studies in Psychology with the statistical analysis techniques currently available? The current studies attempted to address those questions by conducting a study of literature review. It was shown that there was still a big space for improvement for studies on communication particularly in Indonesia. Many statistical techniques that were currently available, such as dyadic analysis or social network analysis, have not been employed to address the issues related to dependence of observation or social relations data. Moreover, statistical models or techniques should be developed to measure communication events to reduce the dependence on self-report assessments.