A Data Science Central Community
From the publisher:
Although many graduate students and researchers have had course work in statistics, they sometimes find themselves stumped in proceeding with a particular data analysis question. In fact, statistics is often taught as a lesson in mathematics as opposed to a strategy for answering questions about the real world, leaving beginning researchers at a loss for how to proceed. In these situations, it is common to turn to a statistical expert, the "go to" person when questions regarding appropriate data analysis emerge. Your Statistical Consultant is an authentic alternative resource for describing, explaining, and making recommendations regarding thorny or confusing statistical issues. Written to be responsive to a wide range of inquiries and levels of expertise, this book is flexibly organized so readers can either read it sequentially or turn directly to the sections that correspond to their concerns and questions. Each chapter opens with a list of questions to be addressed, followed by an overview of the chapter. Key terms are bold-faced within the chapter and key points are italicized. Chapter headings are followed by detailed answers to questions, including conceptual explanations and clarifications of the use and nuances of a particular technique or issue. Examples and opinions of contemporary statistical experts are cited throughout the book.
The book is organized into four parts that identify the covered topics. Part I addresses issues surrounding the acquisition of data and how to ensure the quality and fidelity of data entry. The authors present numerous suggestions and rules of thumb for data entry, designed to avoid the common pitfalls of the novice researcher. Extensive recommendations and guidelines are provided for researchers wishing to collect original data, including via the Internet, and for those wishing to utilize an existing database to conduct secondary analyses. Part II addresses the logic of traditional approaches to statistical analysis and provides an extensive discussion of the challenges to that logic and new paradigms for analysis. It includes guidelines for selecting the appropriate statistical test. The assumptions underlying traditional null hypothesis significance testing are discussed in detail. A new chapter addresses the use of statistical models and illustrates both how to interpret simple models and how to develop a model from a set of hypotheses. The last chapter of Part II provides guidelines for the selection of statistical tests. Part III provides responses to frequently asked questions about data. Chapter 8 discusses the implications of missing data, extreme data (outliers), and data that are not normally distributed, while Chapter 9 discusses issues surrounding the measurement properties of data and the use of dummy variables. Finally, Part IV addresses the two most commonly utilized statistical techniques, analysis of variance (ANOVA) and multiple regression analysis (MRA). These techniques are described and issues of appropriate sample size, statistical interaction, and violations of assumptions are discussed. The concluding chapter of Part IV explores the connections between different statistical techniques, the use of meta-analysis and "modern robust statistics," and offers recommendations for success.