An Introduction to Statistics and Data Analysis Using Stata® From Research Design to Final Report To our parents, Betty, Joe, Ginny, and Steve An Introduction to Statistics and Data Analysis Using Stata® From Research Design to Final Report Lisa Daniels Washington College Nicholas Minot International Food Policy Research Institute For Information: Copyright © 2020 by SAGE Publications, Inc. SAGE Publications, Inc. All rights reserved. Except as permitted by U.S. copyright law, no part of this work may be reproduced or distributed in any form or by any 2455 Teller Road means, or stored in a database or retrieval system, without permission Thousand Oaks, California 91320 in writing from the publisher. E-mail: [email protected] When forms and sample documents appearing in this work are SAGE Publications Ltd. intended for reproduction, they will be marked as such. Reproduction 1 Oliver’s Yard of their use is authorized for educational use by educators, local school sites, and/or noncommercial or nonprofit entities that have purchased 55 City Road the book. London, EC1Y 1SP United Kingdom SAGE Publications India Pvt. Ltd. B 1/I 1 Mohan Cooperative Industrial Area Printed in the United States of America Mathura Road, New Delhi 110 044 India Library of Congress Cataloging-in-Publication Data SAGE Publications Asia-Pacific Pte. Ltd. Names: Daniels, Lisa, author. 18 Cross Street #10-10/11/12 Title: An introduction to statistics and data analysis using China Square Central Stata : from research design to final report / Lisa Daniels, Singapore 048423 Washington College, Nicholas Minot, International Food Policy Research Institute, Washington, DC. Description: First edition. | Thousand Oaks, California : SAGE, [2018] | Includes bibliographical references and index. Acquisitions Editor: Leah Fargotstein Identifiers: LCCN 2018035896 | ISBN 9781506371832 (Paperback : acid-free paper) Editorial Assistant: Claire Laminen Content Development Editor: Chelsea Neve Subjects: LCSH: Stata. | Social sciences–Statistical methods– Production Editor: Karen Wiley Computer programs. | Quantitative research–Computer programs. Copy Editor: QuADS Prepress Pvt. Ltd. Classification: LCC HA32 .D37 2018 | DDC 005.5/5–dc23 LC Typesetter: Integra record available at https://lccn.loc.gov/2018035896 Proofreader: Scott Oney Indexer: William Ragsdale This book is printed on acid-free paper. Cover Designer: Ginkhan Siam Marketing Manager: Shari Countryman 19 20 21 22 23 10 9 8 7 6 5 4 3 2 1 BRIEF CONTENTS Preface xiv Acknowledgments xix PART I • THE RESEARCH PROCESS AND DATA COLLECTION 1 Chapter 1 • The Research Process 2 Chapter 2 • Sampling Techniques 10 Chapter 3 • Questionnaire Design 25 PART II • DESCRIBING DATA 41 Chapter 4 • An Introduction to Stata 42 Chapter 5 • Preparing and Transforming Your Data 59 Chapter 6 • Descriptive Statistics 74 PART III • TESTING HYPOTHESES 109 Chapter 7 • The Normal Distribution 110 Chapter 8 • Testing a Hypothesis About a Single Mean 131 Chapter 9 • Testing a Hypothesis About Two Independent Means 142 Chapter 10 • One-Way Analysis of Variance 157 Chapter 11 • Cross Tabulation and the Chi-Squared Test 172 PART IV • EXPLORING RELATIONSHIPS 185 Chapter 12 • Linear Regression Analysis 186 Chapter 13 • Regression Diagnostics 217 Chapter 14 • Regression Analysis With Categorical Dependent Variables 253 PART V • WRITING A RESEARCH PAPER 283 Chapter 15 • Writing a Research Paper 284 APPENDICES 303 Appendix 1 • Quick Reference Guide to Stata Commands 303 Appendix 2 • Summary of Statistical Tests by Chapter 319 Appendix 3 • Decision Tree for Choosing the Right Statistic 325 Appendix 4 • Decision Rules for Statistical Significance 326 Appendix 5 • Areas Under the Normal Curve (Z Scores) 328 Appendix 6 • Critical Values of the t Distribution 330 Appendix 7 • Stata Code for Random Sampling 332 Appendix 8 • Examples of Nonlinear Functions 338 Appendix 9 • Estimating the Minimum Sample Size 350 Glossary 354 AbouttheAuthors 360 NameIndex 361 SubjectIndex 363 DETAILED CONTENTS Preface xiv Acknowledgments xix PART I •  THE RESEARCH PROCESS AND DATA COLLECTION 1 Chapter1•TheResearchProcess 2 1.1 Introduction 3 1.2 Read the Literature and Identify Gaps or Ways to Extend the Literature 4 1.3 Examine the Theory 6 1.4 Develop Your Research Questions and Hypotheses 6 1.5 Develop Your Research Method 7 1.6 Analyze the Data 8 1.7 Write the Research Paper 8 Exercises 8 References 9 Chapter2•SamplingTechniques 10 2.1 Introduction 11 2.2 Sample Design 12 2.3 Selecting a Sample 14 2.3.1ProbabilityandNonprobabilitySampling 14 2.3.2IdentifyingaSamplingFrame 16 2.3.3DeterminingtheSampleSize 17 2.3.4SampleSelectionMethods 18 2.4 Sampling Weights 21 2.4.1CalculatingSamplingWeights 21 2.4.2UsingSamplingWeights 23 Exercises 23 References 24 Chapter3•QuestionnaireDesign 25 3.1 Introduction 26 3.2 Structured and Semi-Structured Questionnaires 26 3.3 Open- and Closed-Ended Questions 28 3.4 General Guidelines for Questionnaire Design 28 3.5 Designing the Questions 30 3.5.1QuestionOrder 30 3.5.2PhrasingtheQuestions 31 3.6 Recording Responses 34 3.6.1ResponsesintheFormofContinuousVariables 34 3.6.2ResponsesintheFormofCategoricalVariables 35 3.7 Skip Patterns 36 3.8 Ethical Issues 38 Exercises 39 References 40 PART II • DESCRIBING DATA 41 Chapter4•AnIntroductiontoStata 42 4.1 Introduction 43 4.2 Opening Stata and Stata Windows 43 4.2.1ResultsWindow 44 4.2.2ReviewWindow 44 4.2.3CommandWindow 44 4.2.4VariablesWindow 45 4.2.5PropertiesWindow 45 4.3 Working With Existing Data 45 4.4 Entering Your Own Data Into Stata 48 4.5 Using Log Files and Saving Your Work 51 4.6 Getting Help 54 4.6.1HelpCommand 54 4.6.2SearchCommand 54 4.6.3StataWebsite 54 4.6.4UCLA’sInstituteforDigitalResearchandEducationWebsite 55 4.7 Summary of Commands Used in This Chapter 55 Exercises 56 Chapter5•PreparingandTransformingYourData 59 5.1 Introduction 59 5.2 Checking for Outliers 60 5.3 Creating New Variables 63 5.3.1Generate 63 5.3.2UsingOperators 64 5.3.3Recode 64 5.3.4Egen 67 5.4 Missing Values in Stata 69 5.5 Summary of Commands Used in This Chapter 69 Exercises 71 References 73 Chapter6•DescriptiveStatistics 74 6.1 Introduction 75 6.2 Types of Variables and Measurement 75 6.3 D escriptive Statistics for All Types of Variables: Frequency Tables and Modes 77 6.3.1FrequencyTables 77 6.3.2Mode 80 6.4 D escriptive Statistics for Variables Measured as Ordinal, Interval, and Ratio Scales: Median and Percentiles 81 6.4.1Median 81 6.4.2Percentiles 82 6.5 D escriptive Statistics for Continuous Variables: Mean, Variance, Standard Deviation, and Coefficient of Variation 83 6.5.1Mean 84 6.5.2VarianceandStandardDeviation 87 6.5.3CoefficientofVariation 88 6.6 D escriptive Statistics for Categorical Variables Measured on a Nominal or Ordinal Scale: Cross Tabulation 91 6.7 Applying Sampling Weights 94 6.8 Formatting Output for Use in a Document (Word, Google Docs, etc.) 96 6.9 Graphs to describe data 96 6.9.1BarGraphs 96 6.9.2BoxPlots 96 6.9.3Histograms 100 6.9.4PieCharts 101 6.10 Summary of Commands Used in This Chapter 104 Exercises 105 References 107 PART III • TESTING HYPOTHESES 109 Chapter7•TheNormalDistribution 110 7.1 Introduction 111 7.2 The Normal Distribution and Standard Scores 112 7.3 Sampling Distributions and Standard Errors 119 7.4 E xamining the Theory and Identifying the Research Question and Hypothesis 121