STAT Activity 11 Section 3: Advanced Statistical Techniques Sections 1 and 2 have served to prepare you for the understanding of advanced statistical techniques. This is the section you have been waiting for, but it could not come too prematurely. To introduce these concepts without a solid understanding of research, exploratory data analysis, assumptions, and simple statistical techniques would really make your head spin. In any case, the course is 2/3 over and you should at least briefly pause to celebrate what you have learned and how your perseverance has paid off. This section will contain three activities and will cover the following analytical strategies (if it becomes difficult to keep all the techniques you are learning straight, refer to the last page of your text – there is a great table that can help you out): ANCOVA. The Analysis of Covariance technique is a life-saver when you are comparing means between defined groups and have an additional variable (or variables) that you would like to ‘control’ for. An example might be: Are mean productivity scores for three groups of work teams different, when you control for length of time on the job? Or: Are depression scores for young, middle, and older adults different after controlling for health, gender, and social support? Factorial ANOVA. When you have more than one predictor variable a Factorial ANOVA design might be just what you are looking for. These techniques include Two-way repeated-measures ANOVA, Two-way Mixed ANOVA, Three-way independent ANOVA, and so on. For example: Perhaps you are going to design a social support study for people suffering from chronic pain. Your study includes two treatment groups and control group. Further, you have every reason to believe (based on past research and theory) that men and women will respond differently to the treatment groups. A factorial design can handle such complexities. Repeated-Measures. If you are examining multiple groups but the same people belong to each group, you will use a repeated-measures design. For example, instead of randomly assigning people to either Treatment A or Treatment B, if you choose to have all participants in both treatments (of course you would need to consider carry-over effects, practice, and counter balancing, etc.) then you have a repeated-measures design. There are some great advantages to repeated-measures design (key among them the ability to reduce the statistical impact of individual differences). MANOVA. With the tests you have learned this far, we have been constrained by one requirement of one outcome variables. A MANOVA allows for a design in which you have groups being compared on multiple outcome variables. For example, if you are interested in comparing men and women and their psychological health. You may have a number of measures that assess the construct of psychological health: depression, life satisfaction, and well-being. A MANOVA allows you to make this comparison with one elegant analysis. Non-Parametric Tests. Now that you have learned a number of parametric techniques…what do you do if your data do not meet parametric assumptions? Non-parametric tests to the rescue! Tests covered under this category include: Chi square, Wilcoxon rank-sum test, Mann-Whitney tests, Kruskal-Wallis test for independent conditions and Freidman’s ANOVA for related conditions. Once you master these additional techniques (and you are well rested) you will be asked to complete the signature assignment which will give you an opportunity to do research on a set of supplied data. Congratulations on completing this graduate level statistics course. You will now have the core competencies related to statistics that will allow you to more fully glean knowledge from your content courses. Statistics is not like riding a bike – if you stop using it, you lose it. So, please do not skip over the results sections in peer reviewed articles…be sure to use all that you have worked so hard for. When you get to your dissertation, you will be glad that you did! Required Reading: Discovering Statistics Using SPSS: Preface, How to Use This Book, Chapters 11, 12, 13, 15, 16 Self-Tests Smart Alex's Quizzes SPSS Data Sets: Activity7.sav Activity8.sav Activity6a.sav Activty6b.sav Activty6c.sav Activity10.sav Education.sav Gss.sav Optional Resources: Interactive Multiple Choice Questions Flashcards Assignment 11 Signature Assignment 4Signature Assignment For the final activity, please thoroughly answer each of the questions below. Your grade on this activity will be based on accuracy and comprehensiveness. Your paper should be between 3500-4200 words using APA formatting. Review the file education.sav. Using the data contained in this 500 sample data set, synthesize an integrated understanding about education in four different areas. In this assignment, you will need to examine the data, determine the appropriate test method being sure that the conditions required for that method have been met, perform the analysis, then interpret the results. Synthesize your findings into an integrated report. Be sure to support your position with data and the appropriate statistical tests as needed. Locate two peer review journal articles that deal with each question; compare and contrast your findings with the peer review research. Prepare a paper suitable for submission to a non-statistician academic conference on adult education using graphs, tables, and figures as necessary while still maintaining appropriate academic rigor. Place all relevant statistical output in an appendix. 1. What is the relationship, if any between education and gender? Discuss any differences that may exist and describe the characteristics of the sample. 2. What is the relationship, if any, between parental education and the education of the respondent? If a relationship exists, which parent has the strongest effect on the educational level of the respondent? 3. Is there a linear relationship between age and education, and if so, how strong is that relationship? Is it possible to predict educational level based on age? If so, what limitations exist for the developed method? 4. What is the relationship of marital status on education? Do singles or married persons tend to be more highly educated? Your writing should demonstrate thoughtful consideration of the ideas and concepts that are presented in the course and provide new thoughts and insights relating directly to this topic. Responses should reflect doctoral-level writing standards and have no spelling, grammar, or syntax errors. Submit your paper in the Course Work area of the Activity screen. Learning Outcomes: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 Assignment Outcomes Review research methods and basic statistics as they relate to planning, conducting, and interpreting inferential statistics. Develop appropriate null and alternative hypotheses given a research question. Calculate, integrate, and evaluate descriptive statistical analysis. Create, integrate, and evaluate visual displays of data. Apply appropriate statistical tests based on level of measurement. Calculate, interpret, and understand the appropriate use of inferential statistical analysis. Evaluate the results of the analysis. Demonstrate how population, sampling, and statistical power are related to inferential analysis. Analyze the assumptions required for valid inferential tests. Evaluate the difference between parametric and non-parametric data analysis and how to apply the correct statistical procedure. Demonstrate proficiency in the use of SPSS. Demonstrate proficiency in reporting statistical output in APA format. Synthesize various statistical concepts, apply them to analyze data, and integrate them into a consistent theoretical framework. Compare and contrast that position with published peer review research on the same topic.