Posts

Finally we come to a close to discuss the assumptions that our random intercept and slope models rely on.

So far we have simply believed that Stata and R give us unbiased standard errors that allow for accurate statistical inference. In this session we learn that it is not that simple.

In this session we will extent the random intercept model in such ways that slopes of micro-level variables may also vary across clusters.

In this session we will use our new knowledge on how to model random slopes and cross-level interaction to replicate a recent study on symbolic and realistic immigration-threat perceptions.

In this session we will use our new multilevel modeling skills to replicate a recent study on xenophobia.

In this session we will use our new multilevel modeling skills to replicate a recent study on anti-Muslim prejudice.

In this session we will learn why we actually need multilevel modeling. What are clustered data and why does clustering matter?

In this session you will learn about the most basic type of multilevel model: The random intercept model, which allows average levels of the outcome to vary across clusters.

In this second session, we will discuss theories that seek to explain why people hold racist and xenophobic prejudices.

In this session we will discuss two recent articles that use OLS regression to analyze (factorial) survey experiments on xenophobia. We will also replicate one of them based on the Danish ESS '14.