E-Campus.Com Project Evaluation

Group Members: James Jones

When I went into this project, I really expected, based on the complaints that I had received from Richland students, that the Richland bookstore would be one of the most expensive stores around. That did not turn out to be the case at all. In fact, Richland's bookstore was one of the cheapest around when looking at the best price.

However, there are certain problems that arose during the comparison of the data that should be worked out if the project was ever to be repeated.

Additional Data

The main problem with the project was the collection of data on used textbook prices. People recorded the used price, even if there was not a used textbook available. One of the complaints I hear from students is that there are never any used textbooks available. However, the bookstore told the Richland group that the used textbook price was 75% of the new textbook price, so there was a price entered, regardless of whether or not there were actually any used textbooks available.

For the online bookstores, there was no reference made to how long it would take the books to arrive or how much the shipping was. With the Richland bookstore, you have tax to pay. With the online bookstores, you have shipping. Both should be figured into the price to make all comparisons equal.

It would also be nice to collect data on what the bookstores would buy the books back for.

Unnecessary Data

We collected the list textbook price, but we never used it in the analysis. It could be used to see how what bookstores sell their books for compares to the publisher's suggested price, but we didn't do that.

Changes the Next Time

If we were to do this again, it would be nice to collect the information on the shipping, handling, and taxes. It would also be nice to add a flag for whether the textbook was actually available in new or used format and variable for how long it would take to get the books delivered.


I think the project was relevant. Books are something that every college student has to deal with and they all feel that they pay too much for them.

Part of the problem with textbook prices is that the publishers continually press their authors to release new books every couple of years, even though there may not be substantial improvements in the text. They make their money by selling new books, not by the bookstores reselling used ones.

There are other problems. The Richland bookstore claims that they can't get old editions of a textbook when a new one comes out. Personally, I don't believe they try to hard. So, even though the faculty may wish to keep the current textbook, the bookstore makes it very difficult to do so.

Additional Topics

I think it would be interesting to do a comparison of costs between bookstores that sell their books and bookstores that rent their books. I know that when I was at Eastern (back in the dark ages), we rented textbooks and they weren't nearly that expensive because the cost was spread over multiple semesters and multiple students. Students had the option of buying the textbook if they wanted to, but most didn't want to, so it was cheaper to rent. The relevancy for all students would be high since they all have to buy textbooks.

I would like to see an updated database on vehicle information for analysis. The current set of data for cars that is widely available was done in 1982. I would like to see data collected on manufacturer, model, class, fuel economy, weight, transmission type, number of cylinders, engine displacement, horsepower, price, and other variables that might define a car. One could then describe cars by any one of a number of classification variables and test to see if there are differences between them. This would also be a good dataset to have for multiple regression. The relevancy would be high since most people drive a car.

I would like to see a good dataset of baseball information. A player's hits, runs, rbi's, hr's, triples, doubles, strike outs, etc, and their annual salary. The same thing could be done with other sports. One could describe the data by position or by team. Inferential statistics could be performed to see if there are differences between the classifications of player and also determine whether certain characteristics determine the salary (regression). The relevancy is low unless you're interested in sports, and then it's just interesting, but not really relevant.