Mitchell Centre Seminar Series - Nynke Niezink (Carnegie Mellon University), Network Analysis in Statistics Education: Substantive and Methodological Results
| Dates: | 22 October 2025 |
|---|
| Times: | 16:00 - 17:30 |
|---|
| What is it: | Seminar |
|---|
| Organiser: | School of Social Sciences |
|---|
| Speaker: | Nynke Niezink |
|---|
|
|
Progress on substantive research and methodological development often goes hand in hand. In this talk, I will illustrate this in the context of statistics education. While statistics classes are getting larger and larger, feeling a sense of belonging remains fundamental to students’ academic success and well-being. Sense of belonging is inherently relational, but has up to now never been studied from a network perspective. In the first part of the talk, I evaluate how students’ social networks and sense of belonging relate to their academic performance in a large undergraduate statistics course, using a network autocorrelation model. In the second part, I zoom in on one of the descriptive statistics in the study, widely used to evaluate homophily: the E-I index. This is the number of between-group (external) minus the number of within-group (internal) ties divided by the total number of ties in a network. I will demonstrate that E-I index values and statistical tests are easily misinterpreted. Using a classical dataset on the information flow between organizations as an example, I argue the importance of using the correct reference value for E-I index interpretation and show how the choice of null distribution can impact the results of significance tests.
Speaker
Nynke Niezink
Role: Associate Professor of Statistics & Data Science
Organisation: Carnegie Mellon University
Travel and Contact Information
Find event
2.016
Arthur Lewis Building
Manchester
