Question Details

Descriptive Statistical Analysis and Chi-Square
This Case Assignment has two parts.  The first part requires that you provide a descriptive analysis of the population sampled for the Ugandan database. The second part requires that you use Chi-Square to test for independence in the selection of new hires for an organization.  develop a text-based description of a database; and, that you identify questions about the population that you want to answer through descriptive statistics.  Your written description of the sample population in the database is designed to help you place the your case in the proper context for other case assignments in this course.  You will also be required to write an objective for your descriptive analysis of the database.
The first part also requires that you prepare the necessary descriptive statistics, frequency tables, and graphs with SPSS to support and answer your questions about the population described in Part A. (This part may be submitted as late a Module 3 since SPSS is required and some students may encounter delays in receiving it for this course.)
You have been emailed two SPSS databases. The Ugandan Database is required for this Module and for all five case assignments for this course.  It is named Ugandan-SPSS-Data.sav  and is a database of the results of a survey of 1476 Ugandan women concerning HIV/AIDS.   (Please note that this and any other SPSS database can only be opened in SPSS.  If SPSS is installed on your computer just double click on the link to the database and it will spawn SPSS to open.)
The second SPSS database is named Hired-Not Hired.  This database is used to answer the Chi-Square based questions for this Module in Part B. 
Part A
For you to understand the Ugandan database and describe it in the proper context you must first study the demographics of Uganda.  There are several very fundamental reasons why this is necessary. The primary reason is that you need to know how representative the population surveyed is of the Ugandan population.  Think carefully about this question before you answer it.  (At this time you should research the demographics of Uganda through the Web. Use Google or any other search engine to search for Ugandan Demographics.)
Discussion Questions & Exercise
For this case you must review the variables identified for each database and identify the variables which can be used to best describe the study population.  Write two questions concerning each database that you believe may be key to characterizing the subject population.  The variables for each database are as follows:
Ugandan-SPSS-Data.sav   (Please note that a nominal variable is also considered as a categorical variable.  If a nominal variable has only two levels itis considered to be a dichtomous variable.)
Variable  (Type)	Label
percept          Nominal	Perception of Chances of getting AIDS
agegrp          Ordinal	Age 5-year groups
region           Nominal	Region
residenc        Nominal	Type of place of residence
marital          Nominal	Current marital status
religion          Nominal	Religion
marage         Scaled	Age at first marriage
condomus     Nominal	Ever use of condom
regpart         Nominal	Have a regular sex partner other than spouse
watches         Nominal	Watches TV every week
listens           Nominal	Listens to radio every week
avoidway       Nominal	Ways to avoid AIDS
sexage         Scaled	Age at first sexual intercourse
Please note that all assignments and the reporting of statistical results MUST be in APA 6th Ed. Format.
Case Assignment Questions:
Part A --  Descriptive Analysis
1.    Based on the information provided in the tables above, briefly describe the study population from which the Ugandan database was sampled.   Identify the key variables in the database which best describe the characteristics of Ugandan women.  Please identify the area of research in health sciences/public health for which the data contained in the databases may be the most useful.
2.    State an objective that you may use for conducting a descriptive study based on the Ugandan database. (When one is exploring a population there are always several questions that must be asked to help you focus on the key issues or questions between the population from which the sample was taken and the study sample or the database in this case. (Hint: The primary question is usually related to the ability of the researcher to make certain statistical inferences about the study population from the sample.)
3.    Present at least two questions based on the Ugandan  database that can be adequately answered with descriptive statistics.
4.    Prepare a descriptive analysis of the Ugandan database based on your objective stated in Part One, Question #2. Use appropriate tables, charts and graphs with supporting text.  Only include the pertinent information from SPSS in your assignment.  All output used should be relevant to the study objective. (Hint: Remember, you are not a statistician)
Part B -- Chi-Square
Read the following scenario and answer the questions that follow.
A.  Was There a Selection Bias?
Last Friday your General Manager called you into his office for a confidential meeting. He received several calls from Human Resources and an attorney representing a recent applicant for a position complaining that they believed the recent examination for entry level Health Analyst positions discriminated against certain ethnic groups. Your General Manager was very upset and did not know how he could present information on the selection of recent hires which would show whether or not there was a selection bias or if the selection process was discriminatory.
You suggest that a Chi-Square Test would be able to distinguish if there was a significant difference between the number of persons who applied for the position by ethnicity and gender and those who were actually offered a position.  You were instructed to go ahead and use the Chi-Square Test to determine if there was or was not a selection bias in the Health Analyst examination/selection process.
The following Table summarizes the number of persons applying for the Health Analyst position by gender and ethnicity:
Comparison by Ethnicity
Ethnicity	Number Applied	Number Women Hired/Not Hired
by Race/Ethnicity	 
Number Men Hired/Not Hired
by Race/Ethnicity	Number Hired (Observed)	Expected
American Indian	4	0	2	2	0	2	1
Asian	13	0	5	8	0	8	5
Afro-American	17	2	8	1	6	3	6
White	47	6	10	13	18	19	18
Hispanic	21	4	6	4	7	8	8
Other	5	0	4	0	1	0	2
Total	107	12	36	28	31	40	40
 
Comparison by Gender
Gender	Number Applied	Number Hired (Observed)	Expected
Males	59	28	22
Females	48	12	18
Total	107	40	40
Hired-Not Hired database for this case
The above database allows you to run a Chi-Square Test for this Case Assignment in SPSS.  However, the data for this Case Assignment can be found in the  Hired-Not Hired SPSS Data File sent to you as an email attachment.  It can also be downloaded by clicking on the above link.
1.  Write a research question AND a hypothesis for a chi-square statistical test performed in this Module.  The Research Question should be followed by the Null Hypothesis and the Alternative Hypothesis.  Present this in the following format:
    Research Question 1:  ______________________________
            H0: __________________
            HA: __________________
2.  Run Chi-Square on the hiring data to determine if a possible hiring bias occurred in the selection process by ethnicity and gender.  What are the findings/test results from the Chi-Square Test?  Present your findings with the appropriate tables and graphs.
3.  Summarize in several paragraphs your findings to the General Manager with any recommendations if any.
 (Your grade will be based primarily on your interpretation of the SPSS output.  You must write your response for an audience with a professional knowledge  of statistics.  APA 5th Ed.)
Method
To run the Chi-Square Test with SPSS do the following:
1.Open data set and click on Analyze » Descriptive Statistics » Crosstabs.
2.Place one variable in Row window and the other in Column window.
3.Click on Statistics button on bottom of Crosstabs window » Click on Chi Square and Correlations.
4. Click on Cells and check on Expected and Observed » Continue
5. Click on OK and Output will appear.
6. Review and interpret the results.
 
B.  Can Chi-Square Be Used with the Ugandan Database?
1.   Review the Ugandan database and formulate a question that you believe Chi-Square will be able test or answer appropriately. Present your esearch question and the Chi-Square test.  Interpret and summarize your Chi-Square test results.  The Research Question should be followed by the Null Hypothesis and the Alternative Hypothesis.  Present this in the following format:
    Research Question 1:  ______________________________
            H0: __________________
            HA: __________________
The above questions should be completed forwarded to the instructor through CourseNet not later than the end of Module One.  Part Two may be submitted as late as the completion of Module Three.  
Please note that writing a hypothesis for  a chi-square is not as simple as it may seem. You have to consider that it is a test of independence.  Think about this point carefully before writing the hypotheses.