第一编 概念与伦理基础
第1章 行为研究的精神
第2章 探索与辩护的环境
第3章 伦理问题、道德两难问题与指导方针
第二编 因变量的可操作化和可测量化
第4章 测量信度和效度
第5章 观察、判断和复合变量
第6章 问卷法、访谈法和日志法
第三编 研究设计的逻辑
第7章 随机控制的实验和因果推论
第8章 非随机化研究与函数关系
第9章 随机与非随机抽取样本
第四编 数据分析的基本原理
第10章 描述、展示以及探究数据
第11章 相关关系
第12章 统计检验力与效应值
第五编 单因素设计
第13章 用t 检验比较均值
第14章 方差分析与F 检验
第15章 单因素对比分析
第六编 因素设计
第16章 多因素方差分析
第17章 方差分析中的交互作用
第18章 重复测量方差分析
第七编 数据分析的补充问题
第19章 显著性检验和列联表分析
第20章 多元数据分析
第21章 元分析:研究结果的比较和综合
第八编 附录
详细目录
PART I CONCEPTUAL AND ETHICAL FOUNDATIONS
Chapter The Spirit of Behavioral Research
Science and the Search for Knowledge
What Do Behavioral Researchers Really Know?
Social Constructionism
Contextualism/Perspectivism
Evolutionary Epistemology
Peirce’s Four Ways of Knowing
Rhetoric, Perceptibility, and Aesthetics
Limitations of the Four Supports of Conviction
Behavioral Research Defi ned
Three Broad Research Orientations
The Descriptive Research Orientation
The Relational Research Orientation
The Experimental Research Orientation
Empirical Principles as Probabilistic Assertions
Orienting Habits of Good Scientifi c Practice
Chapter Contexts of Discovery and Justifi cation
Inspiration and Explanation
Theories and Hypotheses
Sources of Inspiration and Insight
Serendipity in Behavioral Research
Molding Ideas Into Working Hypotheses
Positivism, Falsifi cationism, and Conventionalism
Type I and Type II Decision Errors
Statistical Signifi cance and the Effect Size
Two Families of Effect Sizes
Interval Estimates Around Effect Sizes
Summing Up
Chapter Ethical Considerations, Dilemmas,and Guidelines
Puzzles and Problems
A Delicate Balancing Act
Historical Context of the American Psychological Association Code
The Belmont Report, Federal Regulations, and the Institutional
Review Board
Principle I: Respect for Persons and Their Autonomy
Principle II: Benefi cence and Nonmalefi cence
Principle III: Justice
Principle IV: Trust
Principle V: Fidelity and Scientifi c Integrity
Costs, Utilities, and Institutional Review Boards
Scientifi c and Societal Responsibilities
PART II OPERATIONALIZATION AND MEASUREMENT OF DEPENDENT VARIABLES
Chapter Reliability and Validity of Measurements
Random and Systematic Error
Assessing Stability and Equivalence
Internal-Consistency Reliability and Spearman-Brown
KR20 and Cronbach’s Alpha
Effective Reliability of Judges
Effective Cost of Judges
Effective Cost of Items
Interrater Agreement and Reliability
Cohen’s Kappa
Replication in Research
Validity Criteria in Assessment
Convergent and Discriminant Validity
Test Validity, Practical Utility, and the Taylor-Russell Tables
Relationship of Validity to Reliability
Chapter Observations, Judgments, and Composite Variables
Observing, Classifying, and Evaluating
Observing While Participating
Maximizing Credibility and Serendipity
Organizing and Sense-Making in Ethnographic Research
Interpreter and Observer Biases
Unobtrusive Observations and Nonreactive Measurements
Selecting the Most Appropriate Judges
Choosing the Number of Response Alternatives
Effects of Guessing and Omissions on Accuracy
Intrinsic Factors and the Level of Accuracy
Applications of Categorical Judgments
Category Scales and Rating Scales
Numerical, Graphic, and Magnitude Ratings
Rating Biases and Their Control
Bipolar Versus Unipolar Scales
Forming Composite Variables
Forming Multiple Composites
Quantifying the Clarity of Composites
Chapter Questionnaires, Interviews, and Diaries
Concerns About Self-Report Data
Open-Ended Versus Structured Items
Critical Incident Technique
Stages in Developing Interview Protocols
Research Interviews by Telephone
Developing Research Questionnaires
Defensiveness, Inconsistency, and Yea-Saying
Cross-Cultural Questionnaire and Interview Research
One-Dimensional and Multidimensional Attitude Scales
Semantic Differentials for Attitudinal Meaning
Q-Sorts for Subjectivity Ratings
Likert Method of Item Analysis
Thurstone Equal-Appearing Intervals Method
Memory and the Use of Self-Recorded Diaries
PART III THE LOGIC OF RESEARCH DESIGNS
Chapter Randomized Controlled Experiments and Causal Inference
Experimentation in Science
Randomized Experimental Designs
Characteristics of Randomization
The Philosophical Puzzle of Causality
Contiguity, Priority, and Constant Conjunction
Four Types of Experimental Control
Mill’s Methods of Agreement and Difference
Between-Group Designs and Mill’s Joint Method
Independent, Dependent, and Moderator Variables
Solomon’s Extended Control Group Design
Threats to Internal Validity
Threats to External Validity
Statistical Conclusion and Construct Validity
Subject and Experimenter Artifacts
Demand Characteristics and Their Control
Interactional Experimenter Effects
Experimenter Expectancy Effects and
Their Control
Concluding Commentary
Chapter Nonrandomized Research and Functional Relationships
Nonrandomized and Quasi-Experimental Studies
Nonequivalent Groups and Historical Controls
Interrupted Time Series and the Autoregressive Integrated Moving Average
Single-Case Experimental Designs
Cross-Lagged Correlational Designs
Invisible Variables and the Mediation Problem
Path Analysis and Causal Inference
The Cohort in Longitudinal Research
Different Forms of Cohort Studies
Subclassifi cation on Propensity Scores
Multiple Confounding Covariates
Chapter Randomly and Nonrandomly Selected Sampling Units
Sampling a Small Part of the Whole
Bias and Instability in Surveys
Simple Random-Sampling Plans
Improving Accuracy in Random Sampling
Confi dence Intervals for Population Estimates
Speaking of Confi dence Intervals
Other Selection Procedures
Nonresponse Bias and Its Control
Studying the Volunteer Subject
Characteristics of the Volunteer Subject
Implications for the Interpretation of
Research Findings
Situational Correlates and the Reduction of
Volunteer Bias
The Problem of Missing Data
Procedures for Dealing With Missing Data
PART IV FUNDAMENTALS OF DATA ANALYSIS
Chapter Describing, Displaying, and Exploring Data
Descriptions of Sampling Units
Frequency Diagrams and Stem-and-Leaf Displays
Box Plots
Comparing Distributions Back to Back
Measures of Central Tendency
Measures of Spread
The Normal Distribution
Standard Scores
Data Not Distributed Normally
Precision of Estimating Population Means
Defi ning Outliers
Coping With Outliers
Exploring the Data
Chapter Correlation
Pearson r
Proportion of Variance Interpretation of Correlation
Binomial Effect-Size Display
Confi dence Intervals for Effect-Size Correlations
Small Correlations, But Important Effects
Counternull Values of Effect Sizes
Spearman Rank Correlation
Ranks as a Transformation
Observations of Disproportionate Infl uence
Point-Biserial Correlation
Exact Tests for Rho
Phi Coeffi cient
Curvilinear (Quadratic) Correlation
Five Product-Moment Correlations
Comparing Correlations
Considering Third Variables
Effects of Variability on Correlations
Chapter Statistical Power and Effect Size Revisited
Why Assess Statistical Power?
The Neglect of Statistical Power
The requivalent Statistic
Cohen’s Multipurpose Power Tables
The t Test for Comparing Two Means
The Signifi cance of a Product-Moment r
Differences Between Correlation Coeffi cients
The Test That a Proportion is .50
The Difference Between Proportions
The Focused Chi-Square Test
F Tests for Focused Comparisons
Additional Strategies for Improving Power
PART V ONE-WAY DESIGNS
Chapter Comparing Means by Standard t Tests
Gosset and the t Test
Two Components of t Tests
Maximizing t
Effect Sizes and Adjustments for Unequal Sample Sizes
Interpreting the Independent Sample t
Computing the Independent Sample t
Reporting the Results
t Tests for Nonindependent Samples
Effect Size and Study Size Components of Nonindependent Sample t
Assumptions Underlying t Tests
Nonparametric Procedures
The Bootstrap, the Jackknife, and Permutation Tests
Chapter Analysis of Variance and the F Test
The F Test and the t Test
The Analysis of “Variances”
Illustration of an Omnibus F
Dividing Up the Total Variance
ANOVA Summary Tables
Distributions of F
After the Omnibus F
Protecting Against “Too Many t Tests”
Bonferroni Procedures
Bonferroni Tolerance Value
Comparing Two Independent Variabilities
Illustration Using Transformations
Comparing Two Correlated Variabilities
Comparing Three or More Independent Variabilities
Comparing Three or More Correlated Variabilities
Summary of Procedures for Comparing Variabilities
Chapter One-Way Contrast Analyses
Focusing Our Questions and Statistical Tests
Contrast F Tests on Original Data
Contrast t Tests on Original Data Carving Contrasts Out of Published Data
Orthogonal Contrasts
Nonorthogonal Contrasts
Effect Size Indices for Contrasts
The BESD and the Binomial Effect-Size Correlation
Overview of the Four Effect-Size Indices
Comparing Competing Contrasts
Combining Competing Contrasts
Optimal Design and the Allocation of Sample Sizes
PART VI FACTORIAL DESIGNS
Chapter Factorial Analysis of Variance
Confronting Persistent Misunderstandings
An Economy of Design
Effects and the Structure of ANOVA
Individual Differences as Error
The Table of Variance
Testing the Grand Mean by t and F
Unweighted Means Analysis for Equal or Unequal Sample Sizes
Effects on F of Unequal Sample Sizes
Unequal Sample Sizes and Contrasts
Transcending Factorial Structure Using Contrasts
Higher Order Factorial Designs
Blocking and the Increase of Power
Blocking and the Analysis of Covariance
Transforming Data to Remove Interactions
Chapter Interaction Effects in Analysis of Variance
The Interpretation of Interaction
Crossed and Uncrossed Combinations of Group Means
Illustration of Mean Polishing
Constructing Tables of Predicted Means
Studying Interactions in Two-Way Tables
Three-Way Factorial Designs
Defi ning Three-Way Interactions
Further Notes on Interpretation
Simplifying Complex Tables of Residuals
Illustration of a Five-Way Interaction
A Note on Complexity
Chapter Repeated Measures in Analysis of Variance
Use of Repeated Observations
Basic Computations
Fixed and Random Effects Error Terms in the Four Basic Combinations
Latin Squares and Counterbalancing
Analysis of Latin Squares
Some Latin Squares May Be Better Than Others
Other Counterbalancing Designs
Three or More Factors
Two Within-Subjects Factors
Aggregating Error Terms
Three Within-Subjects Factors
Fixed or Random Factors
Did Repeated Measures Help?
The Intraclass r Revisited
Composite Variables and Additional Assumptions
Contrasts in Repeated Measures
Hierarchically Nested Designs
The Choice of Error Terms
PART VII ADDITIONAL TOPICS IN DATA ANALYSIS
Chapter Significance Testing and Association in Tables of Counts
Table Analysis and Chi-Square
The - df Chi-Square Test
Larger Tables of Counts
Distributions of Chi-Square
Procedures for Larger Contingency Tables
Subdividing Tables to Test Specifi c Hypotheses
Fisher Exact Probability Test
Strengthening the Fisher Exact Test
Adjustments for in _ Tables
Complete Partitioning of Larger Tables
The Corner-Cells Test Subtable
Contrasts in Proportions
Alternative Analyses for Smaller Sample Studies
Standardizing Row and Column Totals
Odds Ratio, Relative Risk, Risk Difference, and Phi
One-Sample Tables of Counts
Multiple-Choice-Type Data and the Proportion Index
Chapter Multivariate Data Analysis
Background
Redescribing Relationships Within Sets of Variables
Cluster Analysis for Redescription
Principal Components Analysis for Redescription
ros31960_ Improving Interpretation by Rotation
Psychometric Applications of Principal Components Analysis
Alternatives for the Redescription of Variables
Multidimensional Scaling Illustration
Relationships Among Sets of Variables
Chapter Meta-Analysis: Comparing and Combining Research Results
Statistical Analysis of Statistical Analyses
Criticisms of Meta-Analysis
Interpreting Two or More Studies
Comparing Two Signifi cance Levels
Comparing Two Effect-Size Correlations
Combining Two Signifi cance Levels
Combining Two Effect-Size Correlations
Overall Comparisons of Three or More Studies
Focused Comparisons of Three or More Studies
Combining Three or More p Levels
Combining Three or More Effect-Size Correlations
Results That Are Not Independent
The File Drawer Problem
An Eye to Variability
PART VIII APPENDIXES
A List of Numbered Equations
B Statistical Tables
Glossary
References
Subject Index
Name Index