# Group Assignment 1

Topics: Variance, Null hypothesis, Salary Pages: 2 (445 words) Published: April 23, 2013
Managerial Statistics (ENMGT 572)
Group Assignment

You are permitted to use SPSS to solve the following problems unless otherwise stated.

Problem 1: Refer to the data file Executive salary.xls. This files shows annual salary of CEOs for various companies and industries. The random variable of interest for this exercise is Sum97. This is the CEO’s annual salary plus bonuses for the year 1997 (referred to as salary here-in). The purpose of this exercise is to determine if there is a significant difference between mean salaries of executives for various industries. There are 8 industries in this data file. Answer the following questions, assume = .05 level for all tests.

a. Briefly explain (in one or two sentences) what type of test approach you would use to determine if there is a difference in the mean salaries among all industries. Also, how would you set up the null and alternate hypothesis? b. Conduct a test to determine if the mean salaries among all industries are equivalent. Also, if there is a difference briefly explain or identify the variables or categories (treatment) that contribute to this difference (You may use Tukey’s Procedure for answering this part) c. Conduct a test to determine if there exist a difference in the mean executive salary for the following groups of industries: I. Industrial and Basic Materials

II. Industrial and Technology
III. Basic Materials and Technology
Before conducting the test above, perform a test on equivalent population variances for each pair of company type. In a brief paragraph, comment of the conclusions of the test results. d. Determine the upper and lower range salary values for each industry at 90% and 95% confidence intervals e. Perform a test of normality (Chi-Square evaluation) on the data for each of the three industries in part c and comment on your findings. First produce a histogram with about 5 categories for this test and perform the test using a...