Module 4 - Case

Case assignment

Once you have a feel for what descriptive statistics are and why they might be interesting, then you can turn to some practical experience. As we said, we recommend the use of Excel to calculate what you need to calculate. The data that we will use for this exercise are contained in a downloadable spreadsheet; click here to download this spreadsheet . The data themselves are from a set prepared for teaching purposes by SPSS Inc. As they say:

" satisf.sav . This is a hypothetical data file that concerns a satisfaction survey conducted by a retail company at 4 store locations. 582 customers were surveyed in all, and each case represents the responses from a single customer."

The first worksheet holds the actual numerical data; the second worksheet (labelled "labels") contains a basic codebook listing the variables and the codes associated with each. From your earlier work with data, you'll recognize that some of the variables are demographic descriptors of the patrons of the store, while some are measures of their buying behavior and others are measures of attitudes.

Now things get a bit more problematical. You're about ready to start doing some analysis. If you have the ability to download and install a program, then we recommend that you check out XLStatistics (found at http://www.deakin.edu.au/~rodneyc/XLStatistics/ ). It's a very useful front end to Excel that makes it easy to use the various statistical functions, once you spend a little time becoming familiar with the XLStat interface itself. Basically, it's a series of Excel workbooks that you unzip into a directory and then call through a macro-based interface to Excel that the program itself installs. There will be a tutorial on the use of this program forthcoming shortly.

If you have access to Excel, but do not have the ability to use XLStat, then you can still fall back on Excel's statistical capabilities by doing it yourself. You can click here to access a page with a number of tutorials on how to carry out specific procedures in Excel. These are not the greatest tutorials in the world, but they basically do the job. There are also a wide variety of other Excel learning tools accessible on the Internet, if you want to search for yourself. The basic point is that you need to acquire a certain facility with using Excel to perform certain basic functions to complete this exercise. Everything here, by the way, can be computed by hand, if you're so inclined -- but I don't recommend it.

Part one of this assignment calls for you to talk to calculate a variety of descriptive statistics and do some interpretation. Here's what we'd like you to do with the user-satisfaction data that you have:

• Begin by distinguishing between the interval-level, ordinal level, and categorical level variables in these data. By way of hint, there are two interval variables and eight ordinal variables; the rest are categorical. Please list those that fall in each category.
• For the interval-level variables, calculate appropriate descriptive statistics, as described in the readings, and in a few sentences, interpret what these results tell you. Since ordinal variables can be treated as interval-level for purposes of many analyses, calculate appropriate descriptive statistics on these as well, and interpret them also.
• For the categorical variables, prepare appropriate frequency tables for the various categories, and in a sentence per variable, interpret the results; what do they tell you about the population of respondents?

So that's basically descriptive statistics -- a few numbers, and a bunch of interpretation based on those numbers. Simple, isn't it?

Part two of the assignment calls for you to get a little more into the process by calculating and interpreting an inferential statistic. The following references describe the t-test concept and procedure (there are more descriptions in the optional and supplemental readings if you need them, or you can just ask your instructor for assistance):

Trochim, W. (2006). Research methods knowledge base: The T-test. http://www.socialresearchmethods.net/kb/stat_t.php

Student's t-test (N.D.) Retrieved November 15, 2009, from http://www.biology.ed.ac.uk/research/groups/jdeacon/statistics/tress4a.html

Basically, it's a test for determining the degree to which we can be sure that a difference between two groups that we observe in our sample would continue to be observed if we were able to observe the entire population from which the sample was drawn. The dependent variable must be an interval scale variable, while the independent variable is a two-valued categorical variable. The XLStat manual tells you how to conduct this test using that program; likewise, there is a tutorial on this on the tutorials page.

• For this part of the exercise, first test the hypothesis that men are willing to drive a longer distance to the store than are women. The IV here is the gender, while the DV is distance driven. Obtain your results, and interpret them in a couple of sentences.
• Second, pick one or more of the ordinal variables, and use it as a dependent variable in a t-test with any of the dichotomous (i.e., two-valued) categorical variables as independent variable. Obtain your results, and again interpret them.

Well, that's probably enough for now. If you want some more things to do, e-mail your instructor, and he'll give you some suggestions for possible extra credit work. But try to concentrate on doing this part first, and getting it right.

Price: $23.65

Solution: The downloadable solution consists of 14 pages, 965 words and 16 charts.
Deliverable: Word Document