Data I Homework
This is a collaborative exercise to be carried out by
teams. The teams are (a) the students in DSIS, and (b) the students in
Marketing.
Obtain a ready-made quantitative dataset that is
non-trivial in the sense that it is big enough and interesting enough
that an analysis from it could be publishable. (This doesn't mean it
can't be fun. For example, it could consist of sports statistics, such
as the
Roussin NBA dataset.)
There is a lot of flexibility in what the data can be.
For example, it could be a classic cases-by-variables matrix, such as
the well-known GSS dataset. It could also be an affiliations dataset,
such as the membership of individuals in clubs (e.g., the
AOM Division Membership dataset).
The data should be ordinal-scaled (or better), or 1/0 presence/absence
data.
Put the data into an excel file and create a tab for a
code book that explains each variable. Also create a second tab that has
the data in standardized form (i.e., mean 0 and sd 1). Finally, make
another tab that contains a correlation matrix -- the correlations
between all possible pairs of variables.
Please post the data to the class wiki site:
Reference:
|
