SSW Courses
Courses being offered for the 2009 Summer Statistics Workshops include:
Exploratory and Confirmatory Factor AnalysisThe purpose of this introductory training session is to present the rationale for three uses of factor analysis (and especially evaluating the validity of scores) and to present the basic concepts of both exploratory (EFA) and confirmatory (CFA) applications and emphasize the linkages between EFA and CFA. This course is for faculty and graduate students who have some familiarity with factor analysis but who would like a refresher, and for researchers who recognize the important applications of factor analytic methods but have not yet studied them and need an introduction to the basic concepts. The course will draw on Dr. Thompson's 2004 factor analysis book, published by APA. [top] Hierarchical Linear ModelingThis course will provide you with an introduction to the theory and application of hierarchical linear models. Most data in the behavioral sciences has a multilevel structure, such as students nested within classrooms, patients nested within hospitals, participants nested within group treatment conditions and repeated measures nested within individuals. The major goals of this course are to understand the concepts related to hierarchical linear models, to specify your own models and analyze the data using one of the HLM programs, and to interpret the statistical findings to lay persons. This course will use the HLM software program including HLM and SPSS MIXED to perform the statistical analyses. [top] Intermediate StatisticsThe purpose of this training session is to extend knowledge in tool level topics such as graphing variable distributions, using measures of center and spread to summarize distributions, comparing variable distributions, specifying sampling distributions to estimate parameters, and using normal distributions to interpret test results. In addition, several more advanced topics to aid in designing, conducting, analyzing, and presenting your own behavioral research will be explored. Primary focus will be on the topics of correlation, simple linear regression, hypothesis testing in the two sample case, oneway analysis of variance, factorial analysis of variance, simple repeated measures analysis of variance, power, and multiple comparisons. MS Excel and SPSS will be the principal data analysis tools. [top] Item Response Theory
This course will provide an overview of item response theory (IRT) and its application in psychological measurement. We will begin with concepts and assumptions common to nearly all IRT models, such as the item response function, local independence and dimensionality. We will then move to IRT models for binary test items (e.g., items scored pass/fail), covering model specification, estimation and fit evaluation. Next, we will discuss some IRT models for polytomous response formats (e.g., Likert items), again focusing on specification, estimation and fit evaluation. Major applications of IRT in adaptive testing, test construction and item bias analyses will be described. For software, we will alternate between the BILOG and Mplus software programs. [top] Mediation and ModerationThis course is designed to provide an overview of mediation and moderation analysis. It will begin by discussing the distinction between mediators and moderators, then describe estimating and probing interactions in regression models. Next, basic mediation analysis will be described, including assumptions of the single mediator model. From there, we will examine computer intensive tests of mediation, multilevel mediation, and longitudinal mediation. Finally, models that incorporate both moderation and mediation will be briefly discussed. This course will draw from Dr. David MacKinnon’s (2008) recent book Introduction to Statistical Mediation, published by Erlbaum. For software, we will primarily be using SAS and Mplus. [top] Meta AnalysisThis four-hour short course will review basic methods for meta-analysis of research. Topics will include 1. problem finding and selection; 2. literature search procedures using electronic and branching library searches; 3. search limitation strategies; 4. development of variables and coding menus as an interactive process; 5. basic computation and descriptive procedures; 6. homogeneity analysis using HLM and Q-statistics; 7. covariation of effects with categorical and interval study variables; 8. special topics: missing data, multilevel models, aggregating findings from previous meta-analyses. Examples from the presenter's recent meta-analyses will be presented and discussed. [top] Mixed Methods ResearchThis course is an interactive session for new and seasoned researchers that will provide a framework in a step-by-step manner for using quantitative and qualitative research approaches within the same study. The instructor will provide many published examples and illustrate how to conduct mixed methods research using both statistical software such as SPSS and SAS and qualitative software such as NVIVO. Participants will be able to understand the historical underpinnings of mixed methods research, define and explain mixed methods research in its current form, describe the major steps in the mixed methods research process, identify goals and objectives for mixed methods research studies, identify the rationale and purpose for mixing quantitative and qualitative approaches, design mixed methods research questions, describe the role of sampling in the mixed methods research process, compare and contrast several mixed methods research designs, describe several ways of collecting data in mixed methods research studies, conduct mixed methods data analyses, link research questions to mixed methods data analysis techniques, identify how to make quality meta-inferences, explain the major legitimation types in mixed methods research, and understand issues and standards involved in conducting, reporting, and publishing mixed methods research. [top] Multivariate AnalysisThe purpose of this training course is to review the basic rationale for utilizing multivariate statistics and to familiarize participants with the use of MANOVA and descriptive discriminant analysis (DDA), predictive discriminant analysis (PDA) and canonical correlation analysis. The course presumes only familiarity with commonly used univariate methods, including ANOVA and multiple regression, and will feature selected chapters from Dr. Thompson's forthcoming book, Foundations of Multivariate Statistics. [top] Nonparametric StatisticsThis half-day workshop is designed to review and evaluate eleven basic nonparametric statistical tests frequently used in behavioral science research. Two major emphases are (1) elaborating the theoretical rationale for each of these eleven nonparametric test statistics and (2) sharing specific examples indicating how SPSS procedures are used to generate data analysis results for each test. This workshop is especially useful for researchers who plan to conduct surveys that use probability sampling methods and for statistics teachers who wish to integrate nonparametric methods into their statistics courses and quantitative methods programs. [top] Structural Equation ModelingThis course introduces students to the major elements of designing and analyzing data using SEM and includes: review of data requirement for analysis; path analysis; assumptions; model estimation, evaluation and interpretation; confirmatory factor analysis and structural equation modeling analysis. Participants will use AMOS and MPLUS as the primary statistical analysis computer programs. The course will include both lecture and interactive hands-on practice with data supplied by the instructor. [top] |
 |
