The purpose of this book is to provide instruction and guidance on preparing quantitative data sets prior to answering a studyAEs research questions. Preparation may involve data management and manipulation tasks, data organization, structural changes to data files, or conducting preliminary analysis such as examining the scale of a variable, the validity of assumptions or the nature and extent of missing data. The oresultsoe from these essential first steps can also help guide a researcher in selecting the most appropriate statistical tests for his/her study. The book is intended to serve as a supplemental text in statistics or research courses offered in graduate programs in education, counseling, school psychology, behavioral sciences, and social sciences as well as undergraduate programs that contain a heavy emphasis on statistics. The content and issues covered are also beneficial for faculty and researchers who are knowledgeable about research design and able to use a statistical software package, but are unsure of the first steps to take with their data. Increasingly, faculty are forming partnerships with schools, clinics, and other institutions to help them analyze data in their extensive databases. This book can serve as a reference for helping them get existing data files in an appropriate form to run statistical analysis. This book is not a replacement for a statistics textbook. It assumes that readers have some knowledge of basic statistical concepts and use of statistical software, or that they will be learning these concepts and skills concurrently throughout the course. SPSS was chosen to illustrate the preparation, evaluation, and manipulation of data. However, students or researchers who do not use SPSS will benefit from the content since the overall structure and pedagogical approach of the book focuses heavily on the data issues and decisions to be made.
Carol S. Parke, Ph.D. is an Associate Professor in the Department of Educational Foundations and Leadership at Duquesne University, teaching statistics, research, and measurement courses in masters and doctoral programs in the School of Education. She has a Ph.D. in Research Methodology and an M.A. in Mathematics/Statistics from the University of Pittsburgh as well as a B.S. in Secondary Mathematics Education from Indiana University of Pennsylvania. Dr. Parke's 20 years of research ranges from the design and analysis of large-scale, long-term state and national assessment projects to working intimately with teachers in the classroom to show them how to collect and use data to make decisions and real-time, real-world improvement. Her work has appeared in research and practitioner journals in measurement, assessment, mathematics education, and statistics and includes "Using Assessment to Improve Middle-Grades Mathematics Teaching and Learning," a book for teachers on using performance assessment in mathematics classrooms.
Section 1. The Sample Module 1. Checking the Representativeness of a Sample Module 2. Splitting a File, Selecting Cases, Creating Standardized Values and Ranks Section 2. Nature and Distribution of Variables Module 3. Recoding, Counting, and Computing Variables Module 4. Determining the Scale of a Variable Module 5. Identifying and Addressing Outliers Section 3. Model Assumptions Module 6. Evaluating Model Assumptions for Testing Mean Differences Module 7. Evaluating Model Assumptions for Multiple Regression Analysis Section 4. Missing Data Module 8. Determining the Quantity and Nature of Missing Data Module 9. Quantifying Missing Data and Diagnosing its Patterns Section 5. Working with Multiple Data Files Module 10. Merging Files Module 11. Aggregating Data and Restructuring Files Module 12. Identifying a Cohort of Students