Textual Analysis in International Relations: A Practical Guide

Classified in Other subjects

Written at on English with a size of 4.79 KB.

Different Approaches to Textual Data Analysis

1. Narrative Analysis

A narrative is essentially what is depicted and represented through an interview or a document.

2. Grounded Theory

An approach to develop theory inductively from textual data.

3. Discourse Analysis and Content Analysis

These are two distinct but related approaches to analyzing text and discourse.

Main Textual Analysis Methods for International Relations

  • Content Analysis
  • Discourse Analysis
  • Process Tracing

Qualitative Content Analysis

Qualitative content analysis is an economical text analysis procedure. It involves counting and quantifying elements in qualitative data.

Steps of Content Analysis

  1. Data Collection: Create a corpus of data by selecting relevant texts, images, or other data sources.
  2. Content Familiarization: Thoroughly read and examine the collected data to understand its content.
  3. Codebook Development: Create a codebook or coding matrix. Codes can be derived from existing research literature and theory or developed inductively by reviewing the data. Codes can be textual, visual, related to characters, or focused on individuals.
  4. Data Entry and Cleaning: Enter the data into a suitable software program and clean it to ensure accuracy and consistency.
  5. Analysis: Analyze the coded data to identify patterns, themes, and relationships.

Visual Analysis vs. Text Analysis

Visual Analysis: Suitable for media documents such as films, photos, and posters.

Text Analysis: Involves techniques like word count and KWIC (Key-Words-in-Context).

Creating a Codebook and Codes

  1. Definition of Codebook: A comprehensive list of all the codes developed for the analysis.
  2. Number of Codes: The process of developing codes continues until reaching saturation (Glaser and Strauss, 1967), where no new issues or themes emerge from the data.

Validating Codes with 7Rs

An effective code should meet the following criteria:

  1. Relevant: To the research topic.
  2. Represents: The issue well.
  3. Recognized: In the data.
  4. Repeated: In data (within or across texts).
  5. Raised: By participants (if applicable).
  6. Ratified: By others in the research team.
  7. Retrieves: Applicable text segments.

Computer-Based Content Analysis Software

  • QDA Miner
  • ATLAS.ti

Process Tracing

Process tracing involves telling an empirical story systematically, highlighting causal processes within a sequence of events. It aims to establish a causal story linking an explanatory variable to an observed outcome.

Purposes of Process Tracing

  1. Identify and Describe Phenomena: Helps identify and systematically describe novel political and social phenomena.
  2. Evaluate Hypotheses: Assists in evaluating pre-existing explanatory hypotheses and generating new ones. It can be used deductively or inductively.
  3. Understand Causal Mechanisms: Provides insights into causal mechanisms.
  4. Address Statistical Limitations: Offers a complementary approach to address limitations posed by statistical tools for causal inference (Collier, 2011: 824).

Process Tracing Steps

  1. Identify Hypotheses: Clearly state the causal hypotheses to be tested.
  2. Establish Timeline: Create a chronological timeline of events relevant to the research question.
  3. Construct Causal Graph: Develop a causal graph that visually represents the hypothesized causal relationships.
  4. Identify Alternative Explanations: Identify alternative event choices or explanations at each stage of the process.
  5. Identify Counterfactual Outcomes: Consider what might have happened if different choices or events had occurred.

How to Conduct Process Tracing

  1. Set Time Boundaries: Define clear starting and ending points for the analysis.
  2. Example: Tunisia Case Study: Determine the appropriate time frame for examining the Tunisian case.
  3. State Causal Mechanisms: Clearly articulate the causal mechanisms linking events and outcomes.
  4. Gather Evidence: Collect evidence from various sources, such as interviews and document analysis, to support the causal claims.

Entradas relacionadas: