Please read each question carefully and answer it completely. Please fill in the tables and short answer questions, copying information from EXCEL where requested. "Extra work" not explicitly needed can be copied into the end of document in the Appendix. If an answer is incorrect, including work in the Appendix might allow for minimal "partial credit." (Note that although you are not always required to "show work" on the homework assignments, you may be required to do so on exams.)
For question #1 you will use the provided EXCEL data, Unit 1 (assignment data file). Please note this file includes worksheets that describe the data, identify the countries andthe source of the data as well as the actual data. The worksheet with the data itself is labeled with the tab "data." I used the data for a project considering the relationship between conflict (and other variables) and foreign aid.
#1: Use the data provided in the EXCEL file to fill in the following information.
(a) Using the "Descriptive Statistics" command from the Data Analysis add-in, please create a table of descriptive statistics for the variables: "donors" and "average." (Hint: make sure you choose "Summary Statistics" in the command box to generate the necessary output.) Paste nicely formatted tables below:
Briefly explain in your own words what the values for "skewness" and "kurtosis" mean for each of these variables. Be sure to address whether/why they are different and what that means specifically.
(b) Use the data to fill in the following crosstab tables (note that this form is what your textbook calls "one variable categorical and the other variable quantitative.") In the first table please fill in the # of observations for each block; in the second table please fill in relative frequencies for each block.
Table 1: Crosstabulation of conflict and the % of a population that lives in a rural area.
|
Percent of Population Living in a Rural Area
|
|
Conflict
|
0-20%
|
21-40%
|
41-60%
|
61-80%
|
81-100%
|
|
0
|
|
|
|
|
|
|
1
|
|
|
|
|
|
|
2
|
|
|
|
|
|
Table 2: Crosstabulation (relative frequencies) of conflict and the % of a population that lives in a rural area. Please round to three decimal places.
|
Percent of Population Living in a Rural Area
|
|
Conflict
|
0-20%
|
21-40%
|
41-60%
|
61-80%
|
81-100%
|
|
0
|
|
|
|
|
|
|
1
|
|
|
|
|
|
|
2
|
|
|
|
|
|
(c) Use the data to compute conditional means (based on the state of conflict) for the variables given in this table. Round your answers to two decimal places.
|
Conditional Means (by state of conflict)
|
|
Conflict
|
donors
|
total
|
life ex
|
rural
|
|
0
|
|
|
|
|
|
1
|
|
|
|
|
|
2
|
|
|
|
|
Briefly discuss whether this table suggests any relationship between conflict and other variables. It may help to refer to the information in the EXCEL spreadsheet about how each of these is measured. Please discuss three specific results from the table with numbers as evidence.
(d) Using the correlation command in EXCEL [=correl(,)] find the sample correlation for the variables "life ex" and "rural." Report the correlation value and discuss how to interpret its sign and magnitude.
#2: You are a partner at a sales firm in DC and on a committee to decide which of two junior employees will be given a promotion. You have this data about the # of times each has made a sale for each on-site sales call (sales visit) in the last two years:
|
Candidate #1
Year 1 Year 2
|
|
Sale
|
15
|
75
|
|
No Sale
|
25
|
175
|
|
Total Visits
|
40
|
250
|
|
Candidate #2
Year 1 Year 2
|
|
Sale
|
70
|
35
|
|
No Sale
|
130
|
85
|
|
Total Visits
|
200
|
120
|
(a) You prepare the following table to help decide, please round proportions to three decimal places:
|
Proportion of sales (by sales call/visit)
|
|
|
Candidate
#1
|
Candidate
#2
|
|
Year 1
|
|
|
|
Year 2
|
|
|
Based on this table which of the candidates would you support to the promotion? Briefly explain using exact values/numbers in your answer.
(b) Fill in this table and again combining the data for both years, please round proportions to three decimal places. Discuss based on this information which candidate should receive the promotion, again please use numbers as "evidence" for your answer.
|
Aggregate Data (both years)
|
|
|
Candidate
#1
|
Candidate
#2
|
|
Sales
|
|
|
|
No Sales
|
|
|
|
Total Visits
|
|
|
|
|
Candidate #1
|
Candidate #2
|
|
Overall
Proportion Sales
|
|
|
(c) Are your recommendations in (a) and (b) the same or different? How would you explain this? (Please discuss "Simpson's Paradox" in your response.
Appendix: You can paste any work you may want me to consider here. Please make sure you label the question number and provide a short description of the question for which the work is relevant.