BUS 308 Statistics for Managers, Ashford, Week 1 to 5

BUS 308 Week 1 DQ 1 Language
Numbers and measurements are the
language of business.. Organizations look at results, expenses, quality levels,
efficiencies, time, costs, etc. What measures does your department keep track
of ? How are the measures collected, and how are they summarized/described? How
are they used in making decisions? (Note: If you do not have a job where
measures are available to you, ask someone you know for some examples or
conduct outside research on an interest of yours.)
BUS 308 Week 1 DQ 2 Levels
Managers and professionals often pay
more attention to the levels of their measures (means, sums, etc.) than to the
variation in the data (the dispersion or the probability patterns/distributions
that describe the data). For the measures you identified in Discussion 1, why
must dispersion be considered to truly understand what the data is telling us
about what we measure/track? How can we make decisions about outcomes and
results if we do not understand the consistency (variation) of the data? Does
looking at the variation in the data give us a different understanding of
results?
BUS 308 Week 1 Problem Set Week One
Problem Set Week One. All
statistical calculations will use the Employee Salary Data set (in Appendix
section).
Using the Excel Analysis ToolPak
function Descriptive Statistics, generate descriptive statistics for the salary
data. Which variables does this function not work properly for, even though we
have some generated results?
Sort the data by either the variable
G or GEN1 (into males and females) and find the mean and standard deviation for
each gender for the following variables: SAL, COMPA, AGE, SR, and RAISE. Use
Descriptive for one gender and the fx functions (AVERAGE and STDEV) for the
other.
What is the probability distribution
table for:
A randomly selected person being a
male in a specific grade?
A randomly selected person being in
a specific grade?
Find:
The z score for each male salary,
based on the male salary distribution.
The z score for each female salary, based
on the female salary distribution.
Repeat question 4 for COMPA for each
gender.
What conclusions can you make about
the issue of male and female pay equality? Are all of the results
consistent? If not, why not?
For additional assistance with these
calculations reference the Recommended Resources for Week One.
BUS 308 Week 2 DQ 1 t-Tests
t-Tests.
In looking at your business, when
and why would you want to use a one-sample mean test (either z or t) or a two-
sample t-test? Create a null and alternate hypothesis for one of these issues.
How would you use the results?
BUS 308 Week 2 DQ 2 Variation
Variation exists in virtually all
parts of our lives. We often see variation in results in what we spend (utility
costs each month, food costs, business supplies, etc.). Consider the measures
and data you use (in either your personal or job activities). When are
differences (between one time period and another, between different production
lines, etc.) between average or actual results important? How can you or your
department decide whether or not the variation is important? How could using a
mean difference test help?
BUS 308 Week 2 Problem Set Week Two
Problem Set Week Two. Complete the
problems below and submit your work in an Excel document. Be sure to show all
of your work and clearly label all calculations. All statistical calculations
will use the Employee Salary Data set (in Appendix section).
Problems
Is either that male or female salary
equal to the overall mean salary? (Two hypotheses, one-sample tests needed.)
Are male and female average salaries
statistically equal to each other?
Are the male and female compa
average measures equal to each other?
4. If the salary and compa mean
tests in questions 2 and 3 provide different equality results, which would be
more appropriate to use in answering the question about salary equity? Why?
5. What other information would you
like to know to answer the question about salary equity between the genders?
Why?
BUS 308 Week 3 DQ 1 ANOVA
In many ways, comparing multiple
sample means is simply an extension of what we covered last week. What
situations exist where a multiple (more than two) group comparison would be
appropriate? (Note: Situations could relate to your work, home life, social
groups, etc.). Create a null and alternate hypothesis for one of these issues.
What would the results tell you?
BUS 308 Week 3 DQ 2 Effect Size
Several statistical tests have a way
to measure effect size. What is this, and when might you want to use it in
looking at results from these tests on job related data?
BUS 308 Week 3 Problem Set Week
Three
Problem Set Week Three. Complete the
problems below and submit your work in an Excel document. Be sure to show all
of your work and clearly label all calculations. All statistical calculations
will use the Employee Salary Data set (in Appendix section).
1.Is the average salary the same for
each of the grade levels? (Assume equal variance, and use the Analysis ToolPak
function ANOVA.) Set up the data input table/range to use as follows:?Put
all of the salary values for each grade under the appropriate grade label.
2.The factorial ANOVA with only two
variables can be done with the Analysis ToolPak function two-way ANOVA with
replication. Set up a data input table like the following: Grade For each empty
cell, randomly pick a male or female salary from each grade. Interpret the
results. Are the average salaries for each gender (listed as sample) equal? Are
the average salaries for each grade (listed as column) equal?
3.Repeat question 2 for the compa
values. Grade
For each empty cell randomly pick a
male or female compa from each grade. Interpret the results. Are the average
compas for each gender (listed as sample) equal? Are the average compas for
each grade (listed as column) equal?
4.Pick any other variable you are
interested in and do a simple two-way ANOVA without replication. Why did you
pick this variable, and what do the results show?
5.What are your conclusions about
salary equity now?
BUS 308 Week 4 DQ 1 Confidence Intervals
Earlier we discussed issues with
looking at only a single measure to assess job-related results. Looking back at
the data examples you have provided in the previous discussion questions on
this issue, how might adding confidence intervals help managers understand
results better?
BUS 308 Week 4 DQ 2 Chi-Square Tests
Chi-square tests are great to show
if distributions differ or if two variables interact in producing outcomes.
What are some examples of variables that you might want to check using the
chi-square tests? What would these results tell you?
BUS 308 Week 4 Problem Set Week Four
Problem Set Week Four. Let’s look at
some other factors that might influence pay. Complete the problems below and
submit your work in an Excel document. Be sure to show all of your work and
clearly label all calculations. All statistical calculations will use the
Employee Salary Data set (in Appendix section).
Is the probability of having a
graduate degree independent of the grade the employee is in?
Construct a 95% confidence interval
on the mean service for each gender. Do they intersect?
Are males and females distributed
across grades in a similar pattern?
Do 95% confidence intervals on the
mean length of service for each gender intersect?
How do you interpret these results
in light of our equity question?
BUS 308 Week 5 DQ 1 Correlation
At times we can generate a
regression equation to explain outcomes. For example, an employee’s salary can
often be explained by their pay grade, appraisal rating, education level, etc.
What variables might explain or predict an outcome in your department or life?
If you generated a regression equation, how would you interpret it and the
residuals from it?
BUS 308 Week 5 DQ 2 Regression
At times we can generate a
regression equation to explain outcomes. For example, an employee’s salary can
often be explained by their pay grade, appraisal rating, education level, etc.
What variables might explain or predict an outcome in your department or life?
If you generated a regression equation, how would you interpret it and the
residuals from it?
BUS 308 Week 5 Final Paper
Identify
an issue in your life (work place, home, social organization, etc.) where a
statistical analysis could be used to help make a managerial decision. Develop
a sampling plan, an appropriate set of hypotheses, and an inferential
statistical procedure to test them. You do not need to collect any data on this
issue, but you will discuss what a significant statistical test would mean and
how you would relate this result to the real-world issue you identified. Your
paper should be three to five pages in length (excluding the cover and
reference pages). In addition to the text, utilize at least three sources to to
support your points. No abstract is required. Use the following research plan
format to structure the paper:
Step 1: Identification of the
problem
Describe what is known about the situation, why it is a concern, and what we do
not know.
Step 2: Research Question
What exactly do we want our study to find out? This should not be phrased as a
yes/no question.
Step 3: Data collection
What data is needed to answer the question, how will we collect it, and how
will we decide how much we need?
Step 4: Data Analysis
Describe how you would analyze the data. Provide at least one hypothesis test
(null and alternate) and an associated statistical test.
Step 5: Results and Conclusions
Describe how you would interpret the results. For example, what would you
recommend if your null hypothesis was rejected and what would you do if the
null was not rejected?
A quick example: Concern if gender
is impacting employee’s pay. H0: Gender is not related to pay. H1: Gender is
related to pay. Approach: Multiple regression equation to see if gender impacts
pay after considering the legal factors of grade, appraisal, education, etc. If
regression coefficient for gender is significant, will need to create residual
list to see which employees show excessive variation from predicted salaries
when gender is not considered.