Prerequisites: Intro to Networks, Linear Models

### Lecture: 10am –12 pm, online–synchronous

### Instructor: Professor Olga Chyzh,

### Email: olga dot chyzh at utoronto dot ca

### Office Hours: by appointment, https://chyzh.youcanbook.me/

#### Overview and Objectives

Social science data are inherently network data. Individuals are embedded within networks of friendships and professional relations; administrative units influence and are influenced by the nearby units; countries are nested within complex alliance and trade networks.

The course will introduce the inferential tools for analyzing such data, including the Exponential Random Graph models (ERGMS), actor-oriented model of network dynamics (SIENA), Latent Space Models (LSMs), and Local Structure Graph Models (LSGMs). For each model, we will work through the mathematical and theoretical foundations, discuss published social science applications of them, and utilize the models on example datasets.

#### Learning Outcomes

This course is designed as a series of weekly modules that build upon each other. Each module covers one or more state-of-the-art approaches to statistical analysis of network data. For each model covered, the objectives are that students will:

- Develop a firm grasp of the assumptions and formulation of the model.
- Understand how to interpret results.
- Understand how to evaluate the fit and assumptions of the model.
- Use software to apply the model.
- Read published substantive applications of the model.
- Evaluate the potential for application of the model to their own research.

Two overarching objectives are that students will (1) develop an ability to compare the relative merits of the various models covered for a given empirical application and (2) develop a comprehensive sense of the state of the literature on statistical models for social networks.

#### Homework

Students will have a weekly homework typically assigned on Monday and due on Friday. For each homework, students will apply the methods covered that week to a dataset made available by the instructor. Each student will submit their homework as an R-script and a pdf file. Each homework will take 1–3 hours to complete.

#### Coding Sessions

For each method covered we will run through applications in R during class. Students are strongly encouraged to follow along during class and review/run through these examples after class. Students will be provided with data, but may also use their own datasets.

#### Grading Scale

Grades will be assigned based on performance on the homework assignments. Each assignment will be graded (0–no submission, 1–does not demonstrate comprehension of the methodology, 2–adequate comprehension demonstrated, 3–excellent comprehension demonstrated). The final grade will be calculated using the following grading scheme to the sum of homework grades.

A+ | >= 9 |
---|---|

A | >= 8 |

A- | >= 6 |

B+ | >= 4 |

B- | < 4 |