Data Analysis Using Stata, Third Edition

Data Analysis Using Stata, Third Edition is a comprehensive introduction to both statistical methods and Stata. Beginners will learn the logic of data analysis and interpretation and easily become self-sufficient data analysts. Readers already familiar with Stata will find it an enjoyable resource for picking up new tips and tricks. The book is written as a self-study tutorial and organized around examples. It interactively introduces statistical techniques such as data exploration, description, and regression techniques for continuous and binary dependent variables. Step by step, readers move through the entire process of data analysis and in doing so learn the principles of Stata, data manipulation, graphical representation, and programs to automate repetitive tasks. This third edition includes advanced topics, such as factor-variables notation, average marginal effects, standard errors in complex survey, and multiple imputation in a way, that beginners of both data analysis and Stata can understand. Using data from a longitudinal study of private households, the authors provide examples from the social sciences that are relatable to researchers from all disciplines. The examples emphasize good statistical practice and reproducible research. Readers are encouraged to download the companion package of datasets to replicate the examples as they work through the book. Each chapter ends with exercises to consolidate acquired skills.

The First TimeStarting StataSetting up your screenYour first analysisDo-filesExiting Stata Working with Do-FilesFrom interactive work to working with a do-fileDesigning do-filesOrganizing your work The Grammar of StataThe elements of Stata commandsRepeating similar commandsWeights General Comments on the Statistical CommandsRegular statistical commands Estimation commands Creating and Changing VariablesThe commands generate and replaceSpecialized recoding commandsRecoding string variables Recoding date and timeSetting missing valuesLabelsStorage types, or the ghost in the machine Creating and Changing GraphsA primer on graph syntaxGraph typesGraph elementsMultiple graphsSaving and printing graphs Describing and Comparing DistributionsCategories: Few or many?Variables with few categoriesVariables with many categories Statistical Inference Random samples and sampling distributionsDescriptive inferenceCausal inference Introduction to Linear RegressionSimple linear regressionMultiple regressionRegression diagnosticsModel extensionsReporting regression resultsAdvanced techniques Regression Models for Categorical Dependent VariablesThe linear probability modelBasic conceptsLogistic regression with StataLogistic regression diagnosticsLikelihood-ratio testRefined modelsAdvanced techniques Reading and Writing DataThe goal: The data matrixImporting machine-readable dataInputting dataCombining dataSaving and exporting dataHandling big datasets Do-Files for Advanced Users and User-Written ProgramsTwo examples of usageFour programming toolsUser-written Stata commands Around StataResources and informationTaking care of StataAdditional procedures References Author Index Subject Index Exercises appear at the end of each chapter.

Ulrich Kohler is a sociologist at the Social Science Research Center Berlin (WZB). Dr. Kohler is an organizer of the German Stata Users Group meetings. Frauke Kreuter is an associate professorin the Joint Program in Survey Methodology (JPSM)at the University of Maryland–College Park, professor in the Statistics Departmentat the Ludwig-Maximilians-University of Munich, andhead of the Statistical Methods group at the Institute for Employment Research (IAB) in Nuremberg, Germany. Both authors are associate editors of the Stata Journal. They coauthored a German textbook, Datenanalyse mit Stata, which was the predecessor of this book. They used Data Analysis Using Stata to teach several classes and short courses at the University of Mannheim, the University of Konstanz, the Free University of Berlin, and the University of California–Los Angeles.