Use the same business problem/opportunity and research variable you wrote about in Week 3.

For the Week 4 paper, I thought going through the sections of the assignment would be best for explaining expectations.

• Identify the types of descriptive statistics that might be best for summarizing the data, if you were to collect a sample.

In your Week 3 paper, you needed to identify data that you would collect for your research problem. For this Week 4 paper, you need to explain what descriptive statistics would be used in your analysis. As noted in feedback on the Week 2 paper, the results of a descriptive statistics analysis are presented in both the text, in a summary table, and possibly in a chart or graphic.  To meet this requirement, think about what descriptive statistics would be used in your analysis. What measures of central tendency (mean, median, or mode) would be appropriate? What measures of variability?  Would measures of shape (skew, kurtosis) be relevant? 

• Apply the types of inferential statistics that might be best for analyzing the data, if you were to collect a sample.

In the real world, you would discuss what statistical test should be used to analyze your research problem,  For example, if you wished to determine if production is higher for the afternoon or night shift, a two sample t test would be used.  However, hypothesis testing (C9-C11 in Jaggia & Kelly) is not assigned for this class.  Since you do not learn how to do this type of basic statistical analysis, this requirement does not make a great deal of sense. You might explain how probability - empirical probability,z scores, or a discrete probability distribution  - could be used for your analysis, which might be placed with the following sections.

• Analyze the role probability or trend analysis might play in helping address the business problem. 
• Analyze the role that linear regression for trend analysis might play in helping address the business problem.
• Analyze the role that a time series might play in helping address the business problem.

What choice you make depends on the data you have. Your feedback on your Week 3 paper will help you identify the likely type of data.  If you planned to collect cross-sectional data, you may be able to use a simple or multiple regression analysis.  If you have planned to collect time series data (by year, quarter, month, week, day), a trend analysis is possible. Both methods involve regression analysis, but there is a 'trick' to using simple linear regression with time series analysis. (A post is forthcoming.)   Please discuss this subject - how to use regression analysis -  in class this week.