Comparing groups : randomization and bootstrap methods using R / Andrew S. Zieffler, Jeffrey R. Harring, Jeffrey D. Long
- Author:
- Zieffler, Andrew, 1974-
- Published:
- Hoboken, N.J. : Wiley, [2011]
- Copyright Date:
- ©2011
- Physical Description:
- xxxii, 298 pages : illustrations ; 25 cm
- Additional Creators:
- Harring, Jeffrey, 1964-
Long, Jeffrey D., 1964-
- Contents:
- Machine generated contents note: 1.1.Getting Started -- 1.1.1.Windows OS -- 1.1.2.Mac OS -- 1.1.3.Add-On Packages -- 1.2.Arithmetic: R as a Calculator -- 1.3.Computations in R: Functions -- 1.4.Connecting Computations -- 1.4.1.Naming Conventions -- 1.5.Data Structures: Vectors -- 1.5.1.Creating Vectors in R -- 1.5.2.Computation with Vectors -- 1.5.3.Character and Logical Vectors -- 1.6.Getting Help -- 1.7.Alternative Ways to Run R -- 1.8.Extension: Matrices and Matrix Operations -- 1.8.1.Computation with Matrices -- 1.9.Further Reading -- Problems -- 2.1.Tabular Data -- 2.1.1.External Formats for Storing Tabular Data -- 2.2.Data Entry -- 2.2.1.Data Codebooks -- 2.3.Reading Delimited Data into R -- 2.3.1.Identifying the Location of a File -- 2.3.2.Examining the Data in a Text Editor -- 2.3.3.Reading Delimited Separated Data: An Example -- 2.4.Data Structure: Data Frames -- 2.4.1.Examining the Data Read into R -- 2.5.Recording Syntax using Script Files -- 2.5.1.Documentation File -- 2.6.Simple Graphing in R -- 2.6.1.Saving Graphics to Insert into a Word-Processing File -- 2.7.Extension: Logical Expressions and Graphs for Categorical Variables -- 2.7.1.Logical Operators -- 2.7.2.Measurement Level and Analysis -- 2.7.3.Categorical Data -- 2.7.4.Plotting Categorical Data -- 2.8.Further Reading -- Problems -- 3.1.Reading In the Data -- 3.2.Nonparametric Density Estimation -- 3.2.1.Graphically Summarizing the Distribution -- 3.2.2.Histograms -- 3.2.3.Kernel Density Estimators -- 3.2.4.Controlling the Density Estimation -- 3.2.5.Plotting the Estimated Density -- 3.3.Summarizing the Findings -- 3.3.1.Creating a Plot for Publication -- 3.3.2.Writing Up the Results for Publication -- 3.4.Extension: Variability Bands for Kernel Densities -- 3.5.Further Reading -- Problems -- 4.1.Graphically Summarizing the Marginal Distribution -- 4.2.Graphically Summarizing Conditional Distributions -- 4.2.1.Indexing: Accessing Individuals or Subsets -- 4.2.2.Indexing Using a Logical Expression -- 4.2.3.Density Plots of the Conditional Distributions -- 4.2.4.Side-by-Side Box-and-Whiskers Plots -- 4.3.Numerical Summaries of Data: Estimates of the Population Parameters -- 4.3.1.Measuring Central Tendency -- 4.3.2.Measuring Variation -- 4.3.3.Measuring Skewness -- 4.3.4.Kurtosis -- 4.4.Summarizing the Findings -- 4.4.1.Creating a Plot for Publication -- 4.4.2.Using Color -- 4.4.3.Selecting a Color Palette -- 4.5.Extension: Robust Estimation -- 4.5.1.Robust Estimate of Location: The Trimmed Mean -- 4.5.2.Robust Estimate of Variation: The Winsorized Variance -- 4.6.Further Reading -- Problems -- 5.1.Graphing Many Conditional Distributions -- 5.1.1.Panel Plots -- 5.1.2.Side-by-Side Box-and-Whiskers Plots -- 5.2.Numerically Summarizing the Data -- 5.3.Summarizing the Findings -- 5.3.1.Writing Up the Results for Publication -- 5.3.2.Enhancing a Plot with a Line -- 5.4.Examining Distributions Conditional on Multiple Variables -- 5.5.Extension: Conditioning on Continuous Variables -- 5.5.1.Scatterplots of the Conditional Distributions -- 5.6.Further Reading -- Problems -- 6.1.Randomized Experimental Research -- 6.2.Introduction to the Randomization Test -- 6.3.Randomization Tests with Large Samples: Monte Carlo Simulation -- 6.3.1.Rerandomization of the Data -- 6.3.2.Repeating the Randomization Process -- 6.3.3.Generalizing Processes: Functions -- 6.3.4.Repeated Operations on Matrix Rows or Columns -- 6.3.5.Examining the Monte Carlo Distribution and Obtaining the p-Value -- 6.4.Validity of the Inferences and Conclusions Drawn from a Randomization Test -- 6.4.1.Exchangeability -- 6.4.2.Nonexperimental Research: Permutation Tests -- 6.4.3.Nonexperimental, Nongeneralizable Research -- 6.5.Generalization from the Randomization Results -- 6.6.Summarizing the Results for Publication -- 6.7.Extension: Tests of the Variance -- 6.8.Further Reading -- Problems -- 7.1.Educational Achievement of Latino Immigrants -- 7.2.Probability Models: An Interlude -- 7.3.Theoretical Probability Models in R -- 7.4.Parametric Bootstrap Tests -- 7.4.1.Choosing a Probability Model -- 7.4.2.Standardizing the Distribution of Achievement Scores -- 7.5.The Parametric Bootstrap -- 7.5.1.The Parametric Bootstrap: Approximating the Distribution of the Mean Difference -- 7.6.Implementing the Parametric Bootstrap in R -- 7.6.1.Writing a Function to Randomly Generate Data for the boot() Function -- 7.6.2.Writing a Function to Compute a Test Statistic Using the Randomly Generated Data -- 7.6.3.The Bootstrap Distribution of the Mean Difference -- 7.7.Summarizing the Results of the Parametric Bootstrap Test -- 7.8.Nonparametric Bootstrap Tests -- 7.8.1.Using the Nonparametric Bootstrap to Approximate the Distribution of the Mean Difference -- 7.8.2.Implementing the Nonparametric Bootstrap in R -- 7.9.Summarizing the Results for the Nonparametric Bootstrap Test -- 7.10.Bootstrapping Using a Pivot Statistic -- 7.10.1.Student's t-Statistic -- 7.11.Independence Assumption for the Bootstrap Methods -- 7.12.Extension: Testing Functions -- 7.12.1.Ordering a Data Frame -- 7.13.Further Reading -- Problems -- 8.1.The Randomization Test vs. the Bootstrap Test -- 8.2.Philosophical Frameworks of Classical Inference -- 8.2.1.Fisher's Significance Testing -- 8.2.2.Neyman-Pearson Hypothesis Testing -- 8.2.3.p-Values -- 9.1.Educational Achievement Among Latino Immigrants: Example Revisited -- 9.2.Plausible Models to Reproduce the Observed Result -- 9.2.1.Computing the Likelihood of Reproducing the Observed Result -- 9.3.Bootstrapping Using an Alternative Model -- 9.3.1.Using R to Bootstrap under the Alternative Model -- 9.3.2.Using the Bootstrap Distribution to Compute the Interval Limits -- 9.3.3.Historical Interlude: Student's Approximation for the Interval Estimate -- 9.3.4.Studentized Bootstrap Interval -- 9.4.Interpretation of the Interval Estimate -- 9.5.Adjusted Bootstrap Intervals -- 9.6.Standardized Effect Size: Quantifying the Group Differences in a Common Metric -- 9.6.1.Effect Size as Distance-Cohen's δ -- 9.6.2.Robust Distance Measure of Effect -- 9.7.Summarizing the Results -- 9.8.Extension: Bootstrapping the Confidence Envelope for a Q-Q Plot -- 9.9.Confidence Envelopes -- 9.10.Further Reading -- Problems -- 10.1.Matching: Reducing the Likelihood of Nonequivalent Groups -- 10.2.Mathematics Achievement Study Design -- 10.2.1.Exploratory Analysis -- 10.3.Randomization/Permutation Test for Dependent Samples -- 10.3.1.Reshaping the Data -- 10.3.2.Randomization Test Using the Reshaped Data -- 10.4.Effect Size -- 10.5.Summarizing the Results of a Dependent Samples Test for Publication -- 10.6.To Match or Not to Match...That is the Question -- 10.7.Extension: Block Bootstrap -- 10.8.Further Reading -- Problems -- 11.1.Planned Comparisons -- 11.2.Examination of Weight Loss Conditioned on Diet -- 11.2.1.Exploration of Research Question 1 -- 11.2.2.Exploration of Research Question 2 -- 11.2.3.Exploration of Research Question 3 -- 11.3.From Research Questions to Hypotheses -- 11.4.Statistical Contrasts -- 11.4.1.Complex Contrasts -- 11.5.Computing the Estimated Contrasts Using the Observed Data -- 11.6.Testing Contrasts: Randomization Test -- 11.7.Strength of Association: A Measure of Effect -- 11.7.1.Total Sum of Squares -- 11.8.Contrast Sum of Squares -- 11.9.Eta-Squared for Contrasts -- 11.10.Bootstrap Interval for Eta-Squared -- 11.11.Summarizing the Results of a Planned Contrast Test Analysis -- 11.12.Extension: Orthogonal Contrasts -- 11.13.Further Reading -- Problems -- 12.1.Unplanned Comparisons -- 12.2.Examination of Weight Loss Conditioned on Diet -- 12.3.Omnibus Test -- 12.3.1.Statistical Models -- 12.3.2.Postulating a Statistical Model to Fit the Data -- 12.3.3.Fitting a Statistical Model to the Data -- 12.3.4.Partitioning Variation in the Observed Scores -- 12.3.5.Randomization Test for the Omnibus Hypothesis -- 12.4.Group Comparisons After the Omnibus Test -- 12.5.Ensemble-Adjusted p-values -- 12.5.1.False Discovery Rate -- 12.6.Strengths and Limitations of the Four Approaches -- 12.6.1.Planned Comparisons -- 12.6.2.Omnibus Test Followed by Unadjusted Group Comparisons -- 12.6.3.Omnibus Test Followed by Adjusted Group Comparisons -- 12.6.4.Adjusted Group Comparisons without the Omnibus Test -- 12.6.5.Final Thoughts -- 12.7.Summarizing the Results of Unplanned Contrast Tests for Publication -- 12.8.Extension: Plots of the Unplanned Contrasts -- 12.8.1.Simultaneous Intervals -- 12.9.Further Reading -- Problems.
- Summary:
- "This book, written by three behavioral scientists for other behavioral scientists, addresses common issues in statistical analysis for the behavioral and educational sciences. Modern Statistical & Computing Methods for the Behavioral and Educational Sciences using R emphasizes the direct link between scientific research questions and data analysis. Purposeful attention is paid to the integration of design, statistical methodology, and computation to propose answers to specific research questions. Furthermore, practical suggestions for the analysis and presentation of results, in prose, tables and/or figures, are included. Optional sections for each chapter include methodological extensions for readers desiring additional technical details. Rather than focus on mathematical calculations like so many other introductory texts in the behavioral sciences, the authors focus on conceptual explanations and the use of statistical computing. Statistical computing is an integral part of statistical work, and to support student learning in this area, examples using the R computer program are provided throughout the book. Rather than relegate examples to the end of chapters, the authors interweave computer examples with the narrative of the book. Topical coverage includes an introduction to R, data exploration of one variable, data exploration of multivariate data - comparing two groups and many groups, permutation and randomization tests, the independent samples t-Test, the Bootstrap test, interval estimates and effect sizes, power, and dependent samples"--
- Subject(s):
- ISBN:
- 9780470621691 (hardback)
0470621699 (hardback) - Bibliography Note:
- Includes bibliographical references (pages 287-298).
View MARC record | catkey: 7296287