Data Science · Advanced level

Advanced level · Module 4 of 6 · 20 guided hours

Text Data and Open-Ended Responses

Analyse open-ended survey answers, feedback and documents with text methods, including careful use of AI tools.

Learning objectives

  • Clean and summarise text data
  • Code and categorise responses systematically
  • Use topic modelling and AI-assisted coding with checks

Topics

  • text cleaning and tokenisation
  • word frequencies and n-grams
  • coding frames
  • topic modelling
  • AI-assisted classification with human checks
  • privacy in text

Lessons

  1. Lesson 1 Cleaning and Exploring Text

    Prepare text for analysis and find frequent words and phrases.

    tokenisationstop wordsn-gramsspelling variation

  2. Lesson 2 Coding Responses Systematically

    Build a coding frame and apply it consistently, checking agreement between coders.

    coding frameinductive and deductive codesdouble codinginter-coder agreement

  3. Lesson 3 Topic Models and AI-Assisted Coding

    Use automated methods to scale text analysis, and check their output against human coding.

    topic modellingclassificationvalidation sampleerror analysis

  4. Lesson 4 Reporting Text Findings

    Combine counts and quotes into findings that represent people's views fairly.

    prevalenceillustrative quotesminority viewsanonymised quotes

Practical activity

Analyse 2,000 synthetic open-ended feedback messages about a cash programme: clean them, build a coding frame, apply it and compare with an AI-assisted first pass.

Expected outcomes

  • Systematic insight from text
  • Responsible, checked use of AI for coding

Assessment

Text analysis report marked on coding reliability, insight and checks of automated methods.

Taught live by an instructor — online or in person. Free course lessons are for self-study and do not lead to a certificate — see registration options.