Re-Introduction to R

Since analyzing data that was collected by others, such as the ANES, involves data transformations like recoding and indexing, we'll review these processes in order to get back in to the world of R programming.

In ANES 2012, there is a set of questions that measure political efficacy. We'll use this as an example. We'll create an index variable for efficacy and then use it as a dependent variable in a regression model

There are two sets of questions that measure political efficacy, the standard set and a revised set. We'll use the standard set: effic_complicstd, effic_undstd, effic_carestd, effic_saystd.

First we examine the questions to see what agreement or disagreement mean with respect to efficacy. Three of the items are pointing in a negative direction -- agreeing with the item indicates lower efficacy. One item, effic_undstd, is pointed in the positive direction. We'll recode it so all the items are pointed in the same direction. This will also give us an index where a larger score indicates higher efficacy, which will make interpretation of the model easier.

Finally, we'll construct a linear model to try to explain political efficacy.

The code we need is here:
http://www.bcsociology.org/online/files/SOCY7113effic.r

For the group exercise, try to improve on the explanatory power of the model I specified.