IBM SPSS For Introductory Statistics: Use And Interpretation, Fourth .

xPREFACEThis book demonstrates how to produce a variety of statistics that are usually included in basicstatistics courses, plus others (e.g., reliability measures) that are useful for doing research. We tryto describe the use and interpretation of these statistics as much as possible in nontechnical,jargon-free language. In part, to make the text more readable, we have chosen not to cite manyreferences in the text; however, we have provided a short bibliography, “For Further Reading,” ofsome of the books and articles that our students have found useful. We assume that most studentswill use this book in conjunction with a class that has a textbook; it will help you to read moreabout each statistic before doing the assignments.Overview of the ChaptersOur approach in this book is to present how to use and interpret the SPSS statistics program in thecontext of proceeding as if the HSB data were the actual data from your research project.However, before starting the assignments, we have three introductory chapters. The first chapterdescribes research problems, variables, and research questions, and it identifies a number ofspecific research questions related to the HSB data. The goal is to use this computer program as atool to help you answer these research questions. (Appendix B provides some guidelines forphrasing or formatting research questions.) Chapter 2 provides an introduction to data coding,entry, and checking with sample questionnaire data designed for those purposes. We developedChapter 2 because many of you may have little experience with making “messy,” realistic dataready to analyze. Chapter 3 discusses measurement and its relation to the appropriate use ofdescriptive statistics. This chapter also includes a brief review of descriptive statistics.Chapters 4 and 5 provide you with experience doing exploratory data analysis (EDA), basicdescriptive statistics, and data manipulations (e.g., compute and recode) using the high school andbeyond (HSB) data set. These chapters are organized in very much the same way you mightproceed if this were your project. We calculate a variety of descriptive statistics, check certainstatistical assumptions, and make a few data transformations. Much of what is done in these twochapters involves preliminary analyses to get ready to answer the research questions that youmight state in a report. Chapter 5 ends with examples of how you might write about thesedescriptive data in a research report or thesis.Chapter 6 provides a brief overview of research designs (e.g., between groups and withinsubjects). This chapter provides flowcharts and tables useful for selecting an appropriate statistic.Also included is an overview of how to interpret and write about the results of an inferentialstatistic. This section includes not only testing for statistical significance but also a discussion ofeffect size measures and guidelines for interpreting them.Chapters 7 through 10 are designed to answer the several research questions posed in Chapter 1as well as a number of additional questions. Solving the problems in these chapters should giveyou a good idea of the basic statistics that can be computed with this computer program.Hopefully, seeing how the research questions and design lead naturally to the choice of statisticswill become apparent after using this book. In addition, it is our hope that interpreting what youget back from the computer will become clearer after doing these assignments, studying theoutputs, answering the interpretation questions, and doing the extra statistics problems.Our Approach to Research Questions, Measurement, and Selection of StatisticsIn Chapters 1, 3, and 6, our approach is somewhat nontraditional because we have found thatstudents have a great deal of difficulty with some aspects of research and statistics but not others.Most can learn formulas and “crunch” the numbers quite easily and accurately with a calculatoror with a computer. However, many have trouble knowing what statistics to use and how to

IBM SPSS FOR INTRODUCTORY STATISTICSxiinterpret the results. They do not seem to have a “big picture” or see how research design andmeasurement influence data analysis. Part of the problem is inconsistent terminology. We arereminded of Bruce Thompson’s frequently repeated, intentionally facetious remark at his manynational workshops: “We use these different terms to confuse the graduate students.” For thesereasons, we have tried to present a semantically consistent and coherent picture of how researchdesign leads to three basic kinds of research questions (difference, associational, and descriptive)that, in turn, lead to three kinds or groups of statistics with the same names. We realize that theseand other attempts to develop and utilize a consistent framework are both nontraditional andsomewhat of an oversimplification. However, we think the framework and consistency pay off interms of student understanding and ability to actually use statistics to help answer their researchquestions. Instructors who are not persuaded that this framework is useful can skip Chapters 1, 3,and 6 and still have a book that helps their students use and interpret SPSS.Major Changes in This EditionThe major change in this edition is updating the windows and text to SPSS/PASW 18. We havealso attempted to correct any typos in the 3rd edition and clarify some passages. We expanded theappendix about Getting Started with SPSS (Appendix A) to include several useful procedures thatwere not discussed in the body of the text. We have expanded the discussion of effect sizemeasures to include information on risk and odds ratios in Chapter 7. As noted earlier, Chapter 5has been expanded to include how to write about descriptive statistics. In addition, we havemodified the format of the write-up examples to meet the new changes in APA format in the 6thedition (2010) of the Publication Manual of the American Psychological Association. Althoughthis edition was written using version 18, the program is sufficiently similar to prior versions ofthis software that we feel you should be able to use this book with earlier and later versions aswell.Instructional FeaturesSeveral user-friendly features of this book include1. Both words and the key windows that you see when performing the statistical analyses. Thishas been helpful to “visual learners.”2. The complete outputs for the analyses that we have done so you can see what you will get(we have done some editing in SPSS to make the outputs fit better on the pages).3. Callout boxes on the outputs that point out parts of the output to focus on and indicate whatthey mean.4. For each output, a boxed interpretation section that will help you understand the output.5. Chapter 6 provides specially developed flowcharts and tables to help you select anappropriate inferential statistic and interpret statistical significance and effect sizes.This chapter also provides an extended example of how to identify and write a researchproblem, research questions, and a results paragraph.6. For the inferential statistics in Chapters 7–10, an example of how to write about the outputand make a table for a thesis, dissertation, or research paper.7. Interpretation questions for each chapter that stimulate you to think about the informationin the chapter.8. Several Extra Problems at the end of each chapter for you to run with the program.9. Appendix A provides information about how to get started with SPSS and how to use severalcommands not discussed in the chapters.10. Appendix B provides examples of how to write research problems and questions/hypotheses;Appendix C shows how to make tables and figures.11. Answers to the odd numbered interpretation questions are provided in Appendix D.12. Two data sets on a student resource site. These realistic data sets provide you with data tobe used to solve the chapter problems and the Extra Problems using SPSS.

xiiPREFACE13. An Instructor Resource Web site is available to course instructors who request access fromthe publisher. To request access, please visit the book page or the Textbook Resource tabs atwww.psypress.com. It contains aids for teaching the course, including PowerPoint slides, theanswers to the even numbered interpretation questions, and information related to the evennumbered Extra Problems. Researchers who purchase copies for their personal use can accessthe data files by visiting www.psypress.com/ibm-spss-intro-stats.Major Statistical Features of This EditionBased on our experiences using the book with students, feedback from reviewers and other users,and the revisions in policy and best practice specified by the APA Task Force on StatisticalInference (1999) and the 6th Edition of the APA Publication Manual (2010), we have includeddiscussions of1. Effect size. We discuss effect size in each interpretation section to be consistent with therequirements of the revised APA manual. Because this program doesn’t provide effect sizesfor all the demonstrated statistics, we often have to show how to estimate or compute them byhand.2. Writing about outputs. We include examples of how to write about and make APA-typetables from the information in the outputs. We have found the step from interpretation towriting quite difficult for students so we put emphasis on writing research results.3. Data entry and checking. Chapter 2 on data entry, variable labeling, and data checking isbased on a small data set developed for this book. What is special about this is that the dataare displayed as if they were on copies of actual questionnaires answered by participants. Webuilt in problematic responses that require the researcher or data entry person to look forerrors or inconsistencies and to make decisions. We hope this quite realistic task will helpstudents be more sensitive to issues of data checking before doing analyses.4. Descriptive statistics and testing assumptions. In Chapters 4 and 5 we emphasizeexploratory data analysis (EDA), how to test assumptions, and data file management.5. Assumptions. When each inferential statistic is introduced in Chapters 7–10, we have a briefsection about its assumptions and when it is appropriate to select that statistic for the problemor question at hand.6. All the basic descriptive and inferential statistics such as chi-square, correlation, t tests,and one-way ANOVA covered in basic statistics books. Our companion book, Leech et al.,4th ed. (in press), IBM SPSS for Intermediate Statistics: Use and Interpretation, alsopublished by Routledge/Taylor & Francis, is on the “For Further Reading” list at the end ofthis book. We think that you will find it useful if you need more complete examples andinterpretations of complex statistics including but not limited to Cronbach’s alpha, multipleregression, and factorial ANOVA that are introduced briefly in this book.7. Reliability assessment. We present some ways of assessing reliability in the cross-tabulation,correlation, and t test chapters of this book. More emphasis on reliability and testingassumptions is consistent with our strategy of presenting computer analyses that studentswould use in an actual research project.8. Nonparametric statistics. We include the nonparametric tests that are similar to the t tests(Mann–Whitney and Wilcoxon) and single factor ANOVA (Kruskal–Wallis) in appropriatechapters as well as several nonparametric measures of association. This is consistent with theemphasis on checking assumptions because it provides alternative procedures for the studentwhen key assumptions are markedly violated.9. SPSS syntax. We show the syntax along with the outputs because a number of professors andskilled students like seeing and prefer using syntax to produce outputs. How to include SPSSsyntax in the output and to save and reuse it is presented in Appendix A. Use of syntax to

IBM SPSS FOR INTRODUCTORY STATISTICSxiiiwrite commands not otherwise available in SPSS is presented briefly in our companionvolume, Leech et al. (in press).Bullets, Arrows, Bold, and ItalicsTo help you do the problems, we have developed some conventions. We use bullets to indicateactions in SPSS windows that you will take. For example: Highlight gender and math achievement.Click on the arrow to move the variables into the right-hand box.Click on Options to get Fig. 2.16.Check Mean, Std Deviation, Minimum, and Maximum.Click on Continue.Note that the words in italics are variable names and words in bold are words that you will see inthe windows and utilize to produce the desired output. In the text they are spelled and capitalizedas you see them in the windows. Bold is also used to identify key terms when they are introduced,defined, or important to understanding.To access a window from what SPSS calls the Data View (see Chapter 2), the words you will seein the pull down menus are given in bold with arrows between them. For example: Select Analyze Descriptive Statistics Frequencies.(This means pull down the Analyze menu, then slide your cursor down to Descriptive Statisticsand over to Frequencies, and click.)Occasionally, we have used underlines to emphasize critical points or commands.We have tried hard to make this book accurate and clear so that it could be used by students andprofessionals to learn to compute and interpret statistics without the benefit of a class. However,we find that there are always some errors and places that are not totally clear. Thus, we wouldlike for you to help us identify any grammatical or statistical errors and to point out places thatneed to be clarified. Please send suggestions to gmorgan@cahs.colostate.edu.AcknowledgmentsThis SPSS/PASW book is consistent with and could be used as a supplement for Gliner, Morgan,and Leech (2009), Research Methods in Applied Settings: An Integrated Approach to Design andAnalysis (2nd ed.), which provides extended discussions of how to conduct a quantitativeresearch project as well as understand the key concepts. Or this SPSS book could be a supplementfor Morgan, Gliner, and Harmon (2006), Understanding and Evaluating Research in Applied andClinical Settings, which is a shorter book emphasizing reading and evaluating research articlesand statistics. Information about both books can be found at www.psypress.com.Because this book draws heavily on these two research methods texts and on earlier editions ofthis book, we need to acknowledge the important contribution of three current and formercolleagues. We thank Jeff Gliner for allowing us to use material in Chapters 1, 3, and 6. BobHarmon facilitated much of our effort to make statistics and research methods understandable tostudents, clinicians, and other professionals. We hope this book will serve as a memorial to himand the work he supported. Orlando Griego was a co-author of the first edition of this SPSS book;it still shows the imprint of his student-friendly writing style.

xivPREFACEWe would like to acknowledge the assistance of the many students who have used earlierversions of this book and provided helpful suggestions for improvement. We could not havecompleted the task or made it look so good without our technology consultants, Don Quick andIan Gordon, and our word processor, Sophie Nelson. Linda White, Catherine Lamana, and AlanaStewart and several other student workers were key to making figures in earlier versions.Jikyeong Kang, Bill Sears, LaVon Blaesi, Mei-Huei Tsay, and Sheridan Green assisted withclasses and the development of materials for the DOS and earlier Windows versions of theassignments. Lisa Vogel, Don Quick, Andrea Weinberg, Pam Cress, Joan Clay, Laura JensenJames Lyall, Joan Anderson, and Yasmine Andrews wrote or edited parts of earlier editions. Wethank Don Quick and Sophie Nelson for writing appendixes for this edition. Jeff Gliner, JerryVaske, Jim zumBrunnen, Laura Goodwin, James Benedict, Barry Cohen, John Ruscio, TimUrdan, and Steve Knotek provided reviews and suggestions for improving the text. Bob Fetch andRay Yang provided helpful feedback on the readability and user friendliness of the text. Finally,the patience of our spouses (Hildy, Grant, Susan, and Terry) and families enabled us to completethe task without too much family strain.

CHAPTER 1Variables, Research Problems, and QuestionsResearch ProblemsThe research process begins with an issue or problem of interest to the researcher. This researchproblem is a statement that asks about the relationships between two or more variables. 1 Almostall research studies have more than two variables. Appendix B provides templates to help youphrase your research problem, and provides examples from the expanded high school and beyond(HSB) data set that is described in this chapter and used throughout the book.The process of moving from a sense of curiosity, or a feeling that there is an unresolved problemto a clearly defined, researchable problem, can be a complex and long one. That part of theresearch process is beyond the scope of this book, but it is discussed in most books about researchmethods and books about completing a dissertation or thesis.VariablesKey elements in a research problem are the variables. A variable is defined as a characteristic ofthe participants or situation in a given study that has different values. A variable must vary orhave different values in the study. For example, gender can be a variable because it has twovalues, female or male. Age is a variable that can have a large number of values. Type oftreatment/intervention (or type of curriculum) is a variable if there is more than one treatment or atreatment and a control group. The number of days to learn something or to recover from anailment are common measures of the effect of a treatment and, thus, are also potential variables.Similarly, amount of mathematics knowledge can be a variable because it can vary from none to alot.However, even if a characteristic has the potential to be a variable, if it has only one value in aparticular study, it is not a variable; it is a constant. Thus, ethnic group is not a variable if allparticipants in the study are European American. Gender is not a variable if all participants in astudy are female.In quantitative research, variables are defined operationally and are commonly divided intoindependent variables (active or attribute), dependent variables, and extraneous variables.Each of these topics is dealt with briefly in the following sections.Operational Definitions of VariablesAn operational definition describes or defines a variable in terms of the operations or techniquesused to make it happen or measure it. When quantitative researchers describe the variables intheir study, they specify what they mean by demonstrating how they measured the variable.1To help you, we have identified the variable names, labels, and values using italics (e.g., gender andmale) and have put in bold the terms used in the windows and outputs (e.g., Data Editor), and we use boldfor other key terms when they are introduced, defined, or are important to understanding. Underlines areused to focus your attention on critical points or phrases that could be missed. Italics are also used, as iscommonly the case, for emphasizing words and for the titl

Statistics Student Version can do all of the statistics in this book. IBM SPSS Statistics GradPack includes the SPSS Base modules as well as advanced statistics, which enable you to do all the statistics in this book plus those in our IBM SPSS for Intermediate Statistics book (Leech et al., in press) and many others. Goals of This Book