In Week 4 you learned how to conduct tests that determine if there are differences between mean scores for groups. However, many research questions require more complex designs that include the ability to control for confounding variables and/or include multiple independent variables. For example: You are interested in outcomes for three different therapeutic techniques and want to control for severity of illness at the outset. Or: You want to examine attitudes towards a new federal law and believe that political affiliation and gender are relevant factors to consider.
This week you will learn these advanced techniques.
Read Chapters 12 and 13 in the text. It will be to your advantage to have SPSS open on your computer as you work through these chapters. While you are reading consider your area of research interest and when you have seen these more advanced ANOVA models applied. How might you use these analytical strategies in your dissertation research?
Complete the Self-Tests in the chapters. Answers are available at: http://www.sagepub.com/field4e/study/smartalex.htm.
Optional Preparation for this weekï¿½s Activity
After completing the above activities, if you feel you need additional instruction on the concepts covered, please choose any of the following activities that will assist you in mastering the core concepts.
Main Task: Application ï¿½ ANCOVA and Factorial ANOVA
You will submit one Word document for this activity. You will create this Word document by cutting and pasting SPSS output into Word.
This Activity consists of two parts. In the first part, you will utilize an existing dataset to compute a factorial ANOVA. All SPSS output should be pasted into your Word document. In the second part, you will be asked to create a hypothetical ANCOVA output table for variables related to your area of research interest.
Part A. SPSS Activity
The ï¿½Activity 5.savï¿½ file contains a dataset of a researcher interested in finding the best way to educate elementary age children in mathematics. In particular, she believes that 5th grade girls do better in small class sizes while boys excel in larger classes. Through the school district, she has arranged a pilot program in which some classroom sizes are reduced prior to the state-wide mathematics competency assessment. In the dataset, you will find the following variables:
Participant: unique identifier
Gender: Male (M) or Female (F)
Small (1) ï¿½ no more than 10 children
Medium (2) ï¿½ between 11 and 19 children
Large (3) ï¿½ 20 or more children
Score: final score on the statewide competency assessment.
To complete this Activity
1. Exploratory Data Analysis.
a. Perform exploratory data analysis on all variables in the data set. Realizing that you have six groups, be sure that your exploratory analysis is broken down by group. When possible, include appropriate graphs to help illustrate the dataset.
b. Compose a one to two paragraph write up of the data.
c. Create an APA style table that presents descriptive statistics for the sample.
2. Factorial ANOVA. Perform a factorial ANOVA using the ï¿½Activity 5.savï¿½ data set.
a. Is there a main effect of gender? If so, explain the effect. Use post hoc tests when necessary or explain why they are not required in this specific case.
b. Is there a main effect of classroom size? If so, explain the effect. Use post hoc tests when necessary or explain why they are not required in this specific case.
c. Is there an interaction between your two variables? If so, using post hoc tests, describe these differences.
d. Is there support for the researcherï¿½s hypothesis that girls would do better than boys in classrooms with fewer students? Explain your answer.
e. Write up the results in APA style and interpret them. Be sure that you discuss both main effects and the presence/absence of an interaction between the two.
Part B. Applying Analytical Strategies to an Area of Research Interest
3. Briefly restate your research area of interest.
Analysis of Covariance. Using your area of interest, identify one independent and two dependent variables, such that the dependent variables would likely be covariates. Now, assume you conducted an ANCOVA that shows both the independent variable as well as the covariate significantly predicts the dependent variable. Create a mock ANCOVA output table (see SPSS Output 11.3 in your text for an example) that supports the relationship shown above. Report your mock finding APA style.
Submit your document in the Course Work area below the Activity screen.
Learning Outcomes: 2, 3, 4, 5, 6, 10, 11