用方法說故事：探析電腦輔助文本分析工具在框架研究之應用

Abstract

As the slogan ＂let the data speak for itself＂ gains currency, it becomes important to reconsider the role of the researcher. This study tries to explore the intertwined relationships among data, tools and researchers. In this study, we review the literature on agenda-setting research and framing research, with a particular focus on research methods and analysis tools. To investigate the use of different analysis tools to study media/public issues, this article introduces three analysis tools, word-frequency analysis, co-occurrence analysis, and sentiment analysis, in addition to the qualitative frame analysis method. We find that different analysis tools tell different stories and researcher choices are always important across different research periods. We think that if researchers consider the results obtained using different tools, they will be able to devise a more comprehensive research structure. Therefore, in this article we suggest that researchers apply multiple tools to their data during the pilot study phase.

人文社會領域「向運算轉」後，「讓資料說話」成為一股新的風潮，然而資料如何能說故事？本文以電腦輔助處理大量社群媒體文本資料為例，說明在資料蒐集;處理及分析過程中，須經各種文本分析工具的內在邏輯及研究者介入，資料的意義方得浮現。本文使用詞頻分析;詞語共現;情緒分析工具及質性框架分析法等文本分析工具，對同一組社群媒體文本資料集進行研究，結果發現個別分析工具確實為研究者帶來不同洞見，卻也各自存在侷限。因而，本文認為電腦輔助的文本分析方法在資料篩選;分析;詮釋的過程中，皆需研究者介入進行調整，結合多重方法的研究設計則可協助研究者從文本資料中找出意義框架，說出一則好故事。