Linguistics 460: Textual Data Analysis with R
UNC-Chapel Hill Linguistics
Fall 2025
Elliott Moreton

Week 12, 2025.11.06.Th

Topics: Testing hypotheses about a population mean.

Class:

• General caveats about null-hypothesis significance testing (from last time's slides)
• Hypotheses about a population mean: t-tests
  • Assumptions
  • Worked example
  • In-class problem
  • Start on Lab 7, t-tests.

Assignment for 11/13 Tu: Lab 7, comparison of reaction times in a synonymy-choice task

Week 12, 2025.11.04.Tu

Topics: Null-hypothesis significance testing (continued).

Class:

• Another in-class problem involving a chi-squared test for equal proportions (added to last time's slides)
• General caveats about null-hypothesis significance testing (from last time's slides)
• Testing hypotheses about a population mean: t-tests
  • Assumptions
  • Worked example
  • In-class problem
• Presentation dates

Assignment:

• For 11/13 Th: First draft of project is due.

Week 11, 2025.10.30.Th

Topics: Null-hypothesis significance testing

Class:

• Overview of NHST
• Logic of NHST
• Work through an example (binomial test)
• Do an in-class problem (chi-squared test for equal proportions)
• Discuss caveats
• Presentation-day choice lottery

Assignment:

• For 10/31 F: Each group please enter its first, second, and third choice for a presentation date into the Canvas survey (see Course Materials for today; will magically appear at the end of class)
• For 11/4 Tu: Please read Navarro 2017, Chapter 5, Section 6, and Chapter 13, through the end of Section 4
• For 11/4 Tu: Reading Quiz 10.

Week 11, 2025.10.28.Tu

Topics: Confidence intervals

Class:

• Computing a z or t confidence interval for a sample from a normally-distributed population.
• Frequentist interpretation of confidence intervals
• Some misinterpretations of CIs.
• Factors determining the width of CIs. How much influence does the experimenter have?
• What if the population isn't normally-distributed?
• Other kinds of CIs.
• Uses of CIs

Assignment:

• Reading for 10/30 Th: Navarro 2017 Chapter 11, through the end of Section 7
• For 10/30 Th: Reading Quiz 9.
• There will be no lab this week; please work on projects instead.
• Respond to email feedback about initial draft of project proposal.
• Four students are still looking for a project group.
• For 11/13: First draft of project is due.

Week 10, 2025.10.23.Th

Topics: Sampling theory.

Class:

• Start Zoom.
• Statistical inference (finish slides from last time)
• Sample statistics and the sampling distribution of the mean
• The Central Limit Theorem
• Start on Lab 06, sampling distributions and confidence intervals.

Assignment for 10/28 Tu:

• Read Navarro 2017, Ch. 10.5, which will introduce confidence intervals. There is no associated quiz.
• Do Lab 06, sampling distributions and confidence intervals
• Watch for feedback from me about the project proposals.

Week 10, 2025.10.21.Tu

Topics: Inferential statistics.

Class:

• Descriptive statistics (continued from last time):
  • Measures of association
  • Cautionary tale: "Anscombe's Quartet"
• Inferential vs. descriptive statistics
• Frequentist vs. Bayesian inference
• Probability distributions. Examples:
  • Binomial
  • Normal
• Parameters and inference

Assignment for 10/23 Th:

• Read Navarro 2017, Ch. 10, through the end of 10.4.
• Reading Quiz 8

Week 9, 2025.10.14.Tu

Topics: Experimental design. Descriptive statistics.

Class:

• Experimental design
  • Correlational vs. manipulative studies
  • Replication and pseudo-replication
  • Confounds and third-variable effects.
• Descriptive statistics
  • Measures of central tendency
  • Measures of variability
  • (Time permitting:) Relations between two numeric variables
    • Linear relationships
    • Correlation coefficients: Pearson's, Spearman's
    • Cautionary tale: Anscombe's Quartet

Assignment: Read Navarro 2017, Chapter 5, Sections 5.1, 5.2, 5.4, 5.7--5.10, and do Reading Quiz 07 (on Canvas). This is about descriptive statistics. Navarro uses base R rather than tidyverse R, so skip 5.5, which presents base R's alternative to group_by () and summarize ().

Week 8, 2025.10.09.Th

Topics: Research projects.

Class:

• Start Zoom.
• Comments on midterm, and time for questions about content.
• Final projects in this class. (Handout)
• Time permitting: Concrete example of final project.
• Return midterms.

Assignment:

• For 10/13 Tu: Read Ruxton and Colegrave 2011, Ch. 2, through the end of Section 2.4 on p. 20.
• Do Quiz 06 on that reading.
• Before 10/23 Th: Introduce yourself in the Discussions area of Canvas. This is optional, but it may help you find project partners.
• For 10/23 Th: Find two or three partners for your final-project group.
• For 10/23 Th: Submit an initial proposal for the final project.

Week 7, 2025.10.02.Th

MIDTERM

Please remember to put your name on your paper.

Week 7, 2025.09.30.Tu

Review day for midterm

Topics: Midterm review

Class:

• Start Zoom.
• Go over questions about course content using Midterm Syllabus (on Canvas)
• Readings?
• Labs?
• Any unanswered questions from the quizzes?

Week 6, 2025.09.25.Th

Topics: LASSO regression for document classification.

Class:

• Still using the IMDb database in the textdata library. (Can download imdb_jr.csv from the Canvas site)
• Linear models (preview)
• Logistic models for binary responses
  • Fit by maximum likelihood
  • Fit by LASSO
• How does performance compare on the database?
• Midterm syllabus (midterm is 10/02 Th)

Week 6, 2025.09.23.Tu

Topics: Evaluating classifier performance. Changing parameters to optimize performance

Before class: Please install the naivebayes and psych packages.

Class:

• Still using the IMDb database in the textdata library. (Can download imdb_jr.csv from the Canvas site)
• Fitting the Naive Bayes classifierl
• Using the classifier to make predictions
• Evaluating model performance: Cohen's kappa
• Adjusting model parameters to optimize performance
  • Varying document-term-matrix sparsity
  • Choice of weighting schemes: binary term frequency, term frequeny, tf-idf
  • Overfitting

Assignment:

• Read Navarro 2017, Ch. 15 to the end of Section 15.3. This will preview linear regression, the basis for the next classification model we will encounter. (We'll cover linear regression in more detail in the second half of the course.)

Reminder: Midterm on Thursday, October 2. (Midterm syllabus available on Thursday, September 25.)

Week 5, 2025.09.18.Th

Class:

• Start Zoom
• Comments on Lab 04 (Dante): stop words and sentiment
• Continued from last time: Naive Bayes classifiers
• Building and training a Naive Bayes classifier in R
• Making predictions with the classifier
• Start on Lab 05.

Assignments:

• For 09/23 Tu: Lab 05, classifying patents
• For 09/23 Tu: Please install the "glmnet" package, which we will need for LASSO regression next week.

Announcement: The midterm will be on Thursday, October 2. A midterm syllabus will be available on Thursday, September 27.

Week 5, 2025.09.16.Tu

Topics: Document classification using supervised learning. Document-term matrices. Intro to Naive Bayes classifiers.

Before class: Please make sure these packages are installed:

• textdata
• tm

Class:

• Introduction to classification models (slides, Canvas):
  • Document classification
  • Supervised learning
  • Example: Internet Movie Database (IMDb) reviews (Maas et al, 2011)
  • Document-term matrices
  • Naive Bayes classifiers: theory

Assignments:

• For 09/18 Th: Read Navarro 2017, Ch. 17, through the end of Section 17.2, about Bayesian statistics...
• ... and do Quiz 05 on that reading
• For 09/18 Th: Install the "e1071" package.
• For 09/18 Th: Please try to install the "glmnet" package, and let me know in class whether the installation succeeds.

Week 4, 2025.09.11.Th

Topics: Sentiment analysis.

Class:

• Discuss Lab 03 on tidy text
• Finish sentiment-analysis slides and in-class problem from last time
• Start Lab 04 on sentiment analysis
• (Time permitting:) Set up for next week's classification models.

Assignments:

• For 9/16 Tu: Lab 04 on sentiment analysis.

Week 4, 2025.09.09.Tu

Topics: Sentiment analysis.

Before class:

• Load the following libraries: tidyverse, tidytext, janeaustenr, stringr

Class:

• Start Zoom
• Sentiment
• The "bag-of-words" model
• Sentiment lexicons
• Using inner_join () to look text words up in a sentiment lexicon
• Checking the sentiment lexica against each other.

Assignment for 9/11 Th:

• Read Chapter 2 of Silge & Robinson's book
• Do Quiz 04 on that reading

Week 3, 2025.09.04.Th

Topics: "Tidy" data and tidy text.

Before class:

• Open the Silge & Robinson 2017, Ch. 1 reading in a browser so that you can conveniently copy and paste text into R.

Class:

• Start Zoom for 1 student (MS)
• UNDERLING slide
• Finish "tidy data, tidy text" (slides from Tuesday)
• bind_rows(),pivot_wider(),pivot_longer()
• Start on Lab 03, on tidy text

Assignment for 9/09 Tu:

• Lab 03, tidy text

Week 3, 2025.09.02.Tu

Topics: "Tidy" data and tidy text.

Before class:

• Install the "janeaustenr" package, and any dependencies that R asks you to install.
• Same with "gutenbergr".
• Same with "stringr".

Class:

• Tidy data, tidy text (slides).
  • "Tidy" data format
  • Examples from janeaustenr and gutenbergr
  • unnest_tokens(),group_by(),anti_join()
  • bind_rows(),pivot_wider(),pivot_longer()
  We most likely won't get all the way to the end. <--'-->

Assignment for 9/04 Th:

• Read Silge & Robinson (2017) Ch 1, "Tidy text"
• Do Quiz 03.

Week 2, 2025.08.28.Th

Topics: Regular expressions.

Before class:

• Make sure you can find the word-frequency database Excel file that we used last Thursday (8/21) for in-class examples.
• Install the "rvest" package, and any dependencies that R asks you to install. (This will be needed for the lab.)

Class:

• Finish the slides from last time on regular expressions.
• Counting and displaying categorical data (more slides).
• Start on Lab 02, regular expressions.

Assignment for 9/02 Tu:

• Lab 02, regular expressions (pickups vs. sports cars)

Week 2, 2025.08.26.Tu

Topics: Data filtering and transformation. Strings.

Before class:

• Install the "babynames" package, and any dependencies that R asks you to install.

Class:

• Structure of a research project.
• Illustration: liquid co-occurrence in U.S. baby names (Martin, 2007)
• Transforming a data frame in a data pipeline with mutate (), arrange (), filter ()
• Using regular expressions to describe strings

Assignment for 8/28 Th:

• Read Wickham et al. 2023, Ch. 14 ("Strings"), Sections 1, 2, and 5, and Ch. 15 ("Regular expressions"), through the end of Section 15.4.3. (We'll get to group_by () and summarize () next week; don't worry about them now.)
• Do Quiz 02.

Week 1, 2025.08.21.Th

Topics: Data frames and data visualization.

Before class:

• Instructor: Start Zoom
• Students: Download "lemmas.csv" from Canvas: Modules: Course materials

• Comments on: Lab 00, AI use policy
• Data frames (slides)
• Data visualization (slides).
• Start on Lab 01, "Replicating a scatterplot"

Assignment for 8/26 Tu:

• Finish Lab 01, "Replicating a scatterplot".

Week 1, 2025.08.19.Tu

Topics: Course organization. R and RStudio.

Class:

• Course organization
  • Syllabus
  • UNC Honor Code.
• Read script on preparedness in the classroom.
• Installing R and RStudio
• Types of data in R (slides)
• Start on Lab 00, "Hello, World!"

i Assignment for 8/21 Th:

• Read Freeman & Ross 2019, Chapter 5: "Introduction to R".
• Read Wickham et al. 2023, Chapter 1: "Data visualization".
• Do Quiz 01, "Course organization and data visualization" (Canvas, Quizzes)
• Do Lab 00, "Hello, World!" (Canvas, Assignments)

Week 0: Before the semester even starts

Assignment for 8/19 Tu:

• Read Sections 1.4 and 1.5 of Technical Foundations of Informatics, by Michael Freeman and Joel Ross, about setting up your computer for R and RStudio. (Note: This textbook also says things about a course at a different university; please ignore that. For instance, it says students in that course are supposed to be turning in assignments using GitHub. That does not apply to LING 460 at UNC-Chapel Hill.)
• Following those instructions, try to install R and RStudio on your computer.
• Please bring your computer, with R and RStudio on it, to class on the first day.