The first part of the assignment will be for you to select a data set from the list of 15 databases I will supply to you on Webcampus. You should only need the database for the assignment but if you think you may need additional data, say from the Census, you may use that after you check with me as to the compatibility the data. The project will consist of your analyses of the dataset using descriptive statistics, crosstabs and t-tests and writing up your results in a paper. You may also include a chi-square analysis of the crosstabs as well. You will also be giving a short presentation on the results at the end of the semester on the date noted above.
The databases for this assignment will be found on Webcampus under the “Term Paper Project and Databases.” Each database folder on Webcampus will contain:
1. A Database Instruction Sheet (see below for further explanation)
2. An SPSS datafile.
3. The database codebook.
DATABASE INSTRUCTION SHEET
Each dataset will have a specific instruction sheet associated with the data. This instruction sheet will introduce you to the following:
1. The database
2. The instructions will indicate variables for you to use in your analysis (you are not limited to these data and you may create new variables if you wish but you must use all the variables as indicated)
3. The concept behind the data and some background material
4. A suggested methodology including recoding instructions, methods to use on the data
5. Interpretation and analysis questions.
While you must, at minimum, follow the analysis instructions in the sheet and you must also provide an analysis of all the descriptive statistics (Mean, Median, Mode, Frequency, Standard Deviation, Skewness and the Standard Error).
You may use additional variables if you wish for your analysis if you believe it will enhance your analysis. All of the inferential statistics have diagnostic and significance statistics that must be interpreted and reported and as well.
PAPER FORMAT (10 PAGES):
1. Introduction to Research Question and/or Policy Issue
2. Statement of the hypothesis (null hypothesis)
a. This section provides a brief overview of what is known, when it became an issue, who champions the issue and how the issue is generally viewed. This discussion must be cited (you may use some of the language in the instruction sheet provided for this)
b.Statement of the hypothesis. Typically this is stated as is an IF—THEN statement.
3. Scholarly Literature Review
(Please use a minimum of 3 but no more than 5 articles in this section) where you summarize and show the evolution of the idea and how you came to your research question from the literature. This will require you to organize a BRIEF discussion into themes and cluster author/s in the scholarly literature who share your view and identify what is the basis of their support
Identify which database was used; the source/s of the data and in what timeframe the data was collected. Also if you recoded, created new variables or reduced the skewness of your data you would discuss that here as well
This section begins by describing the data in terms of descriptive statistics (i.e. the mean, median, mode, range, skewness, frequency distributions). Frequency intervals can be reported here is there is an interesting distribution of the data. You may use some of the language from the database instruction sheet to write this section.
5. Analysis [This is the core of the assignment]
Here is where you will describe the main inferential techniques that you used and the inferences from the data and describe the interaction between the independent, intervening and dependent variables. This section should include the following:
• What is to be analyzed? [Description, inferences, and models, i.e. the hypothesis to be tested]
• Describe what was anticipated and provide a picture that summarizes what is observed. It is necessary to attach all your output pages but it is better form to create new tables with the numerical results in a professional looking table in the body of the report.
• Indicate the statistical test used – Put SPSS output in a cleaner table using Word or Excel
• Explain the test, why it is appropriate and what it means.
•What can be concluded from this analysis? Should the null hypothesis be accepted or rejected, in full or in part? Why or why not? (use the interpretation and analysis questions at the end of the dataset instruction sheet to guide your thinking here). Talk about future research.
•List all references and cite these references in the text of your paper using the appropriate bibliographic style.
1. Provide a clear, thoughtful, and coherent presentation of your research question/s, analysis and conclusions.
2. Limit your presentation to about 5 minutes.
3. Provide a visual to aid the audience in following your research.
4. Try to limit the number of slides to about 8-10; limit words on slide to 15 or less
5. Do not read presentation, but engage audience.
The grading of the project will occur at the end of the semester although I encourage you to consult with me about your progress on the project throughout the semester. The presentation on your project will occur on the last week of the semester.
ASSIGNMENT MINIMUM REQUIREMENTS
1. Run all the statistical procedures listed in the instruction sheet including descriptive statistics, diagnostic tests significance tests for the t-tests and crosstabs.
2. Use 3-5 journal articles for your literature review
3. Develop 1-3 hypotheses to test using the test. Include all the descriptive statistical procedures listed above and the t-test statistical procedures
4. Include (separately) all SPSS output material for every statistical procedure you ran during the semester (all that you used in the paper and those not included).
5. Paper (10 pages) with professional looking tables in the body of the paper. NO SPSS OUTPUT IN THE BODY OF THE PAPER
6. Presentation (NO SPSS OUTPUT IN THE PRESENTATION).
This is an individual project so NO GROUP PROJECTS WILL BE ALLOWED. Because there are not enough databases to go around for the class, those students who have chosen the same database as another student must do a slightly different project from the other student using the database. The database instruction sheet indicates several suggested research questions. Students sharing the same database must confer with the other student and decide which research questions they will use for their paper and let me know what they decide via email.
COMPARISON OF MEANS TESTS: T-TEST AND ANOVA REFERENCE GUIDE
You may use the t-test that makes the most sense given your data. Here is a short list of the types of t-tests you should consider. You will most likely not use the one-sample t-test or the ANOVA but I provide it hear as a reference and a contrast to the other two types of tests.
One-Sample t-test: Tests whether the mean of a single variable differs from a specified constant. The assumptions include the population follows normal distribution.
Independent Sample t-test: Helps you to compare the means for two groups. The assumption is each population follows a normal distribution. The variances of two populations may be constant. If this is the case, pooled variance can be used, in order to have a better estimate of the common variance. Otherwise, non-constant variance t-test is more appropriate. The assumption of normality can be checked using Q-Q plot. The assumption of constant variance requires a separate F-test for comparing variances.
Paired Sample t-test: Compares the means of a variable observed at two different situations of a single group. The two situations are often two different times (Before and After a certain treatment). For example, a researcher may be interested in the effect of weight loss of a weight control program. Individuals in the experiment will be measured for various responses such as weight before and after taking the weight control program training. Two weights are measured from the same experimental unit BEFORE and AFTER the treatment. The effect of the weight loss can be examined using Paired t-test.
SPSS Options- This allows users to change the Confidence Interval from the default value of 95%. Options also let you determine how you will handle missing values.
One-Way ANOVA: Analysis of variance for a quantitative dependent variable of a single factor. This is an extension of independent t-test when the number of levels of the factor is more than two. Typical assumptions are normality and constant variance for each level. The normality assumption can be checked using Q-Q plot or some normality test statistics. The constant variance can also be examined by using 'Homogeneity of Variance Tests' such as Levene's test in SPSS. The SPSS submenus are:
Contrasts- This allows you to partition the between-groups sum of squares into trend components (polynomial). Alternatively, one can specify contrasts, which are determined prior to the data collection..
Post Hoc- Here you can choose a test and use it to determine which means differ. Tukey is generally used if you have a large number of comparisons. For a small number of comparisons, Bonferroni can also be used.
Options- You can get descriptive statistics for each level of factor variable and test for homogeneity of variance.
|Due By (Pacific Time)||12/03/2013 12:00 am|
out of 1971 reviews
out of 766 reviews
out of 1164 reviews
out of 721 reviews
out of 1600 reviews
out of 770 reviews
out of 766 reviews
out of 680 reviews