Statistics 6590
Consulting Projects
Due Date: December 9, 2002

You have been asked to take on the role of a statistical consultant, and provide analysis of a dataset that has been supplied to you. This handout will give you some idea of what I shall be looking for in your written projects.

You should break up your project into the following sections. As you will see, some sections should be written in such a way that they will be easily understood by the client that supplied your data. Other sections will be more statistical, so I can assess the quality of the statistical analysis you have done, and to ensure that you understand the statistical methodology for the work you are doing.

You have the choice of either typing your project or writing it by hand, although typed is preferred.

Format of Report

In preparing your project, break your report up into the following sections, and use these sections titles to aid in making your project clear and easy to follow:

• Introduction: Include a description of data, along with the questions of interest you will address in your analysis, or the goals of the study. This section should be written in language that the client will understand.

• Exploratory Analysis: In this section you should include things like plots, tables or summary statistics (means, medians, etc.) Not all of these may be needed, or be applicable, for your dataset. This section should also avoid complicated statistical language, and you should discuss what this exploratory work tells us about the variables, and what questions the client wants to answer.

• Analysis I: Include descriptions of any models used, confidence intervals, hypothesis testing results (including hypotheses, test statistic values and p-values), and other appropriate details of your models and inference. Make sure any conclusions you make are phrased in the context of the problem being addressed.

  Address any assumptions made or possibly violated. If there are assumptions that appear to be violated, you should attempt any data transformation, etc. that may be appropriate.

  This section would normally be much more technical than your client would desire. However, it will show me that you understand the statistical methodology that you are using.

• Analysis II: This will be a summary of the results you give in Analysis I, but written in a way that the client will be able to understand. Therefore, you will not use technical words like null hypothesis or test statistic. You may want to summarize some of your results in a table, similar to the tables we've seen in the consulting papers we've studied.

  This section will be shorter than Analysis I.

• Conclusions: A brief summary of findings, including any limitations you feel there are on how the results can be generalized. You should avoid complex statistical terms in this section.

• Appendix: Containing any computer output.