Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/138530
Title: Argument and information structures in sociology research papers : analysis of the abstract and introduction sections
Authors: Cheng, Wei-Ning
Keywords: Library and information science::Knowledge management
Social sciences::Sociology
Issue Date: 2020
Publisher: Nanyang Technological University
Source: Cheng, W.- N. (2020). Argument and information structures in sociology research papers: Analysis of the abstract and introduction sections. Doctoral thesis, Nanyang Technological University, Singapore. .
Abstract: This is a study of the argument structure and information structure of research papers in the field of sociology. The study adopted an interdisciplinary approach: it addressed the issue of information use in research report writing, within the broader field of information behavior; it applied techniques of ontology and knowledge representation from information science to ana-lyze information structures in research papers; it adopted rhetorical and argument functions from genre studies (a sub-area of applied linguistics) to analyze argument structure; finally it identified links between information structure and argument structure. The objectives of the study were to investigate the argument structure and information structure of the Abstract and Introduction sections of sociology research articles, the relation between argument structure and information structure, as well as the differences across types of sociology research. Based on a sample of 70 research articles from ten sociology journals with high impact factors, the study identified five types of sociology research (i.e. Investigative re-search, Development and evaluation research, Descriptive research, Historical analysis, and Identification research), three main types of argument strategy (i.e. Generic, Literature-focused, and Theory-focused argument strategies), and six information use strategies (i.e. De-scribing information, Elaborating information, Applying information, Constructing new infor-mation, Converting own knowledge to information and Being influenced by information). Due to the small sample sizes for Historical analysis and Identification research, these articles were combined with Descriptive research for the purpose of analysis in this study. For the argument structure analysis, the study developed a coding scheme comprising 43 types of argument claims and supporting arguments, and 13 non-argument elements (i.e. de-scriptive information and context information). After tagging text chunks (mainly sentences) with argument elements, sequential pattern mining was performed to identify textual argument patterns, that is, common sequences of argument elements in the text, and logical argument steps, which are argument support => claim pairs, irrespective of sequence order. For information structure analysis, a Research Information Model was developed com-prising a set of semantic frames, which included: • Three semantic frames associated with particular research types: Research-relation frame associated with Investigative research, Development and Evaluation frame asso-ciated with Development and evaluation research, and Descriptive frame associated with Descriptive research. • Three other semantic frames which model common information structures in all types of research: Comparison frame, Theory/model/framework frame, and Measurement frame. This study also sought to investigate the relation between argument and information structure by analyzing how information is used to link a supporting argument to an argument claim. The analysis results for the Abstracts indicated that the Research objective and Research result are the core argument elements in the Abstracts. Across the types of research, no significant differences in the frequencies of argument elements were found. However, comparing the three argument strategies, the Literature-focused argument strategy was associated with a higher frequency of Topic centrality claims in the Abstracts. From the analysis of textual argument patterns, it was found that each type of research has its own basic textual argument pattern in the Abstract, suggesting that sociology researchers doing a particular type of research have a common understanding of the prototypical argument structure in the research paper. The analysis results for the Introduction sections supported the notion that the Research objective claim is the core claim in the Introduction. Significant differences in the argument profile were found across the research types and the three argument strategies. The analysis of logical argument steps found that researchers generally supported the Research objective with Research gap, Research motivation/justification and Topic centrality across the research types. Within the Abstract and Introduction sections, multiple semantic frames were usually instantiated, reflecting different types of information and different structures/patterns. Multiple semantic frames linked together were analyzed as link patterns. Different types of research also specify some types of information more often than others. Looking at how information is used in logical argument steps, the Converting own knowledge to information strategy is usually used in the supporting argument for a Research gap claim, indicating what is known about a particular concept in the support. The Describing information strategy is often used in the supporting argument for the Research objective claim, to indicate a particular concept in the support directly and simply. This study makes contributions to three disciplines: • For applied linguistics (especially for genre studies and English for Academic purposes), this study identified (1) three sub-genres of academic writing, related to types of re-search as well as the information and argument profiles for each research type, (2) three main types of argument strategy with their argument profiles, (3) the coding scheme for argument structure, and (4) research methods for identifying textual argument patterns as well as for analyzing semantic relations within and across argument claims (with im-plications for argument mining). • For information science, this study identified (1) the Research Information Model that is used to represent important patterns of information, and the links between those pat-terns in research papers, (2) the information profile of research papers for each research type, and (3) relations between information and argument structures. • For the information behavior field (especially for information use), this study defined six information use strategies in the academic context to show how a concept is applied in support of a claim.
URI: https://hdl.handle.net/10356/138530
DOI: 10.32657/10356/138530
Rights: This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:WKWSCI Theses

Files in This Item:
File Description SizeFormat 
DoctoralThesis_Wei-NingCheng.pdfWei-Ning Cheng's doctoral thesis5.05 MBAdobe PDFThumbnail
View/Open

Page view(s) 50

458
Updated on Jan 26, 2023

Download(s) 50

68
Updated on Jan 26, 2023

Google ScholarTM

Check

Altmetric


Plumx

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.