Here is a link to a data set involving the relationship between the diversity of children's and caregivers' speech. The data are simulated in such a way that they are very similar to a real data set (Huttenlocher, Waterfall, Vasilyeva, Vevea & Hedges, 2010). The first column represents a measure of the diversity of children's speech. The second column is the same variable for the caregivers, measured concurrently. The third column is the raw number of speech tokens uttered by the caregivers.
Your assignment this week is deceptively simple. At this point, you know quite a bit about regression, both with several predictors and with a single predictor. Here we have a data set in which a characteristic of children's speech can potentially be predicted by one or both of two measures of caregivers' speech. Use what you have learned about regression to conduct an analysis of the data, with attention to interpretation. You will want to investigate simple (one-predictor) and multiple regressions, investigate the degree to which regression assumptions are satisfied or violated, consider remedies if assumptions are violated, consider whether collinearity is a potential problem, investigate the question of whether outlying or influential observations have distorted your analyses, and so on. Write a report of your investigation, supporting it with appropriate graphical and statistical information.