: Jorge González, Marie Wiberg
: Applying Test Equating Methods Using R
: Springer-Verlag
: 9783319518244
: 1
: CHF 132.90
:
: Bildungswesen
: English
: 217
: Wasserzeichen/DRM
: PC/MAC/eReader/Tablet
: PDF
This book describes how to use test equating methods in practice. The non-commercial software R is used throughout the book to illustrate how to perform different equating methods when scores data are collected under different data collection designs, such as equivalent groups design, single group design, counterbalanced design and non equivalent groups with anchor test design. The R packages equate, kequate and SNSequate, among others, are used to practically illustrate the different methods, while simulated and real data sets illustrate how the methods are conducted with the program R. The book covers traditional equating methods including, mean and linear equating, frequency estimation equating and chain equating, as well as modern equating methods such as kernel equating, local equating and combinations of these. It also offers chapters on observed and true score item response theory equating and discusses recent developments within the equating field. More specifically it covers the issue of including covariates within the equating process, the use of different kernels and ways of selecting bandwidths in kernel equating, and the Bayesian nonparametric estimation of equating functions. It also illustrates how to evaluate equating in practice using simulation and different equating specific measures such as the standard error of equating, percent relative error, different that matters and others.
Foreword7
References9
Preface10
References12
Contents13
Acronyms18
List of Symbols20
1 General Equating Theory Background24
1.1 Introduction24
1.1.1 A Conceptual Description of Equating25
1.1.2 A Statistical Model View of Equating25
1.2 Statistical Models26
1.2.1 General Definition, Notation, and Examples26
1.2.2 Types of Statistical Models27
1.2.3 Mathematical Statistics Formulation of the Equating Problem29
1.2.4 Mathematical Form of the Equating Transformation30
1.2.5 Continuization31
1.2.6 Requirements for Comparability of Scores32
1.2.7 Assessing the Uncertainty of Equating Results32
1.3 Collecting Data in Equating33
1.3.1 Data Collection Designs in Equating34
1.3.1.1 Single Group Design34
1.3.1.2 Equivalent Groups Design34
1.3.1.3 Counterbalanced Design34
1.3.1.4 Non Equivalent Groups with Anchor Test Design35
1.3.1.5 Non Equivalent Groups with Covariates Design35
1.4 Some Examples of Equating Transformations36
1.4.1 The Equipercentile Equating Function36
1.4.2 The Linear Equating Function37
1.4.3 The Kernel Equating Function37
1.5 R Packages That Are Used in This Book38
1.6 Summary and Overview of the Book38
References39
2 Preparing Score Distributions42
2.1 Data42
2.1.1 Data from Ch2:kolenbrennan201442
2.1.2 Data from Ch2:vondavieretal200443
2.1.3 The ADM Admissions Test Data43
2.1.4 The SEPA Test Data44
2.2 Preparing the Score Data44
2.2.1 Functions to Create Score Frequency Distributions45
2.2.2 Score Data in the EG Design45
2.2.3 Score Data in the SG Design50
2.2.4 Score Data in the NEAT Design53
2.3 Presmoothing the Score Distributions56
2.3.1 Polynomial Log-Linear Models for Presmoothing56
2.3.2 Polynomial Log-Linear Smoothing in equate58
2.3.3 Examples59
2.3.3.1 Smoothing Univariate Distributions59
2.3.3.2 Smoothing a Bivariate Distribution60
2.3.4 Choosing the Best Log-Linear Model61
2.4 Using Other Arguments, Packages and Functions64
2.5 Summary65
References65
3 Traditional Equating Methods66
3.1 Equipercentile, Linear, and Mean Equating Transformations66
3.2 Assumptions in the Different Designs67
3.2.1 Assumptions in EG, SG, and CB Designs67
3.2.2 Assumptions in the NEAT Design68
3.3 Traditional Equating Methods for the EG, SG and CB Designs69
3.4 Traditional Equating Methods for the NEAT Design69
3.4.1 Linear Equating Methods for the NEAT Design70
3.4.1.1 Tucker Equating70
3.4.1.2 Nominal Weights Equating71
3.4.1.3 Levine Observed-Score Equating71
3.4.1.4 Levine True-Score Equating72
3.4.1.5 Chained Linear Equating73
3.4.2 Equipercentile Equating Methods for the NEAT Design73
3.4.2.1 Frequency Estimation73
3.4.2.2 Chained Equipercentile Equating74
3.4.2.3 Braun-Holland Equating74
3.5 Examples with the equate Function75
3.5.1 The equate Function75
3.5.2 Examples Under the EG and SG Designs76
3.5.3 Examples Under the NEAT Design83
3.5.3.1 Linear Methods83
3.5.3.2 Equipercentile Methods83
3.5.3.3 Comparison Between Linear and Equipercentile Methods84
3.5.4 Examples Using the ADM Data Under the NEAT Design86
3.6 Additional Features in equate86
3.7 Performing Traditional Equating Methods with SNSequate87
3.8 Comparing Traditional Test Equating Methods88
3.8.1 Bootstrap Standard Errors of Equating88
3.8.2 Bias and RMSE89
3.8.3 Examples Using equate90
3.8.4 Additional Example: A Comparison of Traditional Equating Methods91
3.9 Summary94
References94
4 Kernel Equating96
4.1 A Quick Overview of Kernel Equating96
4.2 Step 1: Presmoothing97
4.2.1 Presmoothing with SNSequate97
4.2.1.1 Presmoothing Under the EG Design98
4.2.1.2 Presmoothing Under the SG Design99
4.2.1.3 Presmoothing Under the CB Design101
4.2.1.4 Presmoothing Under the NEAT Design101
4.2.1.5 Modeling Complexities in the Data102
4.2.2 Presmoothing with kequate104
4.2.2.1 Presmoothing Under the EG Design104
4.2.2.2 Presmoothing Under the SG Design105
4.2.2.3 Presmoothing Under the CB Design106