Saturday, November 15, 2008
HW Due November 18th
Conduct three bivariate regressions using the same continuous outcome, but three different continuous or dichotomous predictors. To what extent does the explanatory power of your predictors differ? Be sure to discuss your findings in terms of how a one-unit change in X influences Y. Importantly, in terms of each of your predictors, what is a “unit”? Select one predictor and create a standard regression table.
Subscribe to:
Post Comments (Atom)
3 comments:
Hi -
A couple of comments regarding some common email questions -
-For the correlations last week, you didn't need to standardize/zscore any measures because it didn't affect the interpretation of the correlation coefficient. But for regression, you should go back to standardizing measures with unfamiliar scales.
-Don't worry too much about skewness and logging variables for now because you haven't learned yet how to do this for regression.
- Only do one table for one predictor.
-A couple of people have asked about 'aveabl'. aveabl is a variable that Doug created and added on to the dataset. It's
a school-level variable - school-average math and literacy ability (a mean aggregate of c1rscale and c1mscale).
Hi Megan,
Could you remind me of how Doug wants to see very small values represented in tables (so small they are in scientific notation)?
It seems to me he didn't want to see scientific notation in the tables, but I may be confused about that. Many thanks,
Karen
Hi Karen
You should round it. So if it's less than 0.005, you would just write it as 0.00
-Megan
Post a Comment