The first thing you have to do is chose a topic, collect data, and then estimate relationships using the techniques of Chapters of multiple regression and regression analysis: model building.

1. Do not use any of the data sets that come with different textbooks. Too many other people have used those data sets. The data source should to be one that you have actually found and verified.

2. Make sure you explain what you are doing clearly with references. The project/paper will have to be properly written. Make sure you following proper citation rules for any information you got from another source, Only APA style.

3. The sample size should of at least five independent variables. The independent variables should not be part of the dependent variable by construction.

4. The project should involve applying the regression analysis in excel only.

5. Somewhere the project you should have information on relevant descriptive statistics.  Relevant does not mean everything Excel can print out.  At a minimum it should include the mean and standard deviation of the variables involved.  However, you may decide that other statistics might provide useful information (e.g. in some cases, the range of the variables might be relevant).  This information could either be on a separate table of descriptive statistics or combined with regression tables.

6. For the regression tables, as mentioned in the information posted on Moodle, you should not just copy from Excel but instead type proper tables.  You should have a regression table that has both your original model (before you eliminate variables) and your final Model.  The sample projects contain some examples.  Note, you do not need separate tables for the two models.

 
NOTE: REGRESSIONS SHOULD BE RUN IN EXCEL ONLY