Using the dataset (11) World Values Survey Scaling A, and the Word doc. copy of the accompanying survey, assume (usually wrongly, but more on that later) that a scale about volunteering might be applicable cross-culturally. Then, assess variables 28-36 ("Do you volunteer for or belong to a church," "do you volunteer or belong to a union," etc.) for all countries in the dataset to answer the following questions with a 2 page report (include a discussion of your methods):

° Who are more "socially involved," Christians (note the recode you'll have to do), Hindus, or those with No Religious Denomination?

° Men or women?

° Is social involvement significantly related to "happiness?"

Some notes - The first thing you have to do is to throw all the "socially involved" questions into a reliability analysis, and get a "correlation matrix" as your output. Normally, this would tell you what items (and how many) "hang together" and that you can therefore scale together. You should still do this, but if you only come up with 2 values that are between (absolute values of) .35 - .7, then don't just create a scale of 2 variables (that's too little; a minimum of 3 is what you want), go ahead and scale all of them together (I think there are 9 items total). I can't check this from this computer... if you find 3 or more that "hang together," then go ahead and scale just those 3 (or however many).

So then you have a scale of "involvement." Right away you know you can test male/ female, or Christian/ Hindu (once a recode is done) against your scale, through a t-test. You can test "country" against the newly created scale through an ANOVA.

Price: $11.59

Solution: The downloadable solution consists of 8 pages, 359 words and 11 charts.
Deliverable: Word Document