LIN 392 Analyzing Linguistic Data: Schedule

Courses‎ > ‎LIN 392 Analyzing linguistic data‎ > ‎

LIN 392 Analyzing Linguistic Data: Schedule

Home

This schedule is subject to change.

Assignments are due at the end of their due date (midnight).

Readings can be done either before or after class; they are chosen to support the material covered in class.

Readings: "ALD" is Baayen's "Analyzing Linguistic Data," "SE" is Hinton's "Statistics Explained."

Week 1, Jan 16

Class canceled because of inclement weather.

Week 2, Jan 23

Introduction

Google trends, Google n-gram viewer
LL breakfast experiments: passive voice, Analyzing real estate listings, and more of the same

First steps in R
- Data: Word counts in inaugural speeches

Please install the R statistics package on your laptop before this lecture. I recommend RStudio. Please use the free version, not the commercial license.
Readings: ALD ch. 1

Week 3, Jan 30

Descriptive statistics: central tendency and spread. Data visualization

Readings: ALD ch. 2
Readings: SE chapter 2
R worksheet: descriptive statistics
R worksheet: plotting
R worksheet: contingency tables
R worksheets: merge and aggregate
Data for exploration: meaning annotation (zipped archive), Fisher word counts

Week 4, February 6

Descriptive statistics: We look again at the worksheets:

Probability distributions, samples and populations

See documents on Canvas: Sampling, and the central limits theorem

Statistical tests: the basic principle; significance thresholds
- Readings: SE ch. 3, 4,5 ALD ch. 3 pp. 44-63
Homework 1 due Thursday February 8 end of day

Week 5, February 13

The t-test

R code: the t-test
R code: reading, preprocessing and counting text.
Readings: SE ch. 6 - 9
Homework 2 due on Thursday February 15 end of day

Week 6, February 20

Statistical testing: t-test exercises, chi-squared. Comparing more than two data sets: ANOVA

Readings: SE ch. 10-12, 19
R code: ANOVA
Data: Fisher word counts, Fisher telephone corpus, analyzed using LIWC (warning: big dataset!), meta-data for the Fisher telephone corpus
Pitfalls of statistical analysis: The problem of multiple testing and the famous Green Jelly Bean story: https://xkcd.com/882/
Discussion in class about your project ideas
Homework 3 due on Thursday February 22 end of day

Week 7, February 27

Correlation and linear regression

Readings: SE ch. 20
R code: Correlation and linear regression
R code: linear regression
A linear regression tutorial
Project proposal due on Thursday Mar 1 end of day

Week 8, March 6

Guest lectures:

2pm: Scott Myers
3:30: Danny Law

Week 9, March 13: Spring Break

Week 10, March 20

Linear regression in R; multiple regression, and different types of predictors

R code: more linear regression
Readings: ALD ch. 4 pp. 84 - 101; SE ch. 21
How not to do a scientific study: "p-value hacking"

Week 11, March 27

Logistic regression

Readings: ALD ch 6 pp 195-199, 202-203
R code: logistic regression

Thursday March 29: Homework 4 due

Week 12, April 3:

Logistic regression and model comparison

Readings: ALD ch. 6 pp 174-188
R code: model comparison

Progress report due

Week 13, April 10

Mixed-effects models in R; practicing regression

Readings: ALD ch. 7 pp 241-284
A mixed-effects regression tutorial with associated data
Model criticism: See this tutorial on linear regression

Thursday April 12: Homework 5 due

Week 14, April 17

More mixed-effects models in R

Model criticism: R worksheet: influential datapoints
See the mixed effects worksheet available on Canvas.

Week 15, April 24

Clustering for data exploration

Readings: ALD ch. 5 up to p. 148
R code: clustering
Topic modeling: a demo
For the topic modeling demo, we use two datasets available on Canvas: r8-train-data and r8-train-meta

Week 16, May 1

Project presentations

Final report due: May 10

Comments

Katrin Erk's homepage	Search this site