Mixed-Effects Regression Models in Linguistics (eBook)

Dirk Speelman is associate professor at the department of linguistics at the KU Leuven. Dirk's main research interest lies in the fields of corpus linguistics, computational lexicology and variational linguistics in general. Much of his work focuses on methodology and on the application of statistical and other quantitative methods to the study of language.  Kris Heylen is a research fellow at the research group Quantitative Lexicology and Variational Linguistics at the University of Leuven (KU Leuven, Belgium) and research fellow at the Institute for the Dutch Language (INT, Leiden, The Netherlands). He specialises in the corpus-based, statistical modelling of lexical semantics and lexical variation.  Dirk Geeraerts is professor of linguistics at the University of Leuven, where founded the research unit Quantitative Lexicology and Variational Linguistics. His main research interests involve the overlapping fields of lexical semantics and lexicology, with a specific descriptive interest in social variation, a strong methodological commitment to corpus analysis, and a theoretical background in Cognitive Linguistics.

Preface 5
Contents 6
1 Introduction 7
1 Mixed Models 7
2 Mixed Models in Linguistics 8
3 Mixed Models in This Book 10
4 Software Used in the Book 11
5 Chapters in This Book 11
References 15
2 Mixed Models with Emphasis on Large Data Sets 17
1 Introduction 18
2 Mixed Models 19
2.1 Linear Mixed Models 19
2.2 Generalized Linear Mixed Models 21
2.3 Estimation and Inference 22
3 Mixed Models in Action: The Leuven Diabetes Project 23
3.1 Interpretation of the Fixed Effects 23
3.2 Tests for Variance Components 25
3.3 Empirical Bayes Estimation 26
4 Issues with Large Data Sets 29
4.1 The Split-Sample Idea 29
4.2 Examples of How Large Data Sets Can Be Split 31
5 Concluding Remarks 32
References 33
3 The L2 Impact on Learning L3 Dutch: The L2 Distance Effect 35
1 Introduction 35
2 Background 37
2.1 CCREMs with Interrelated Random Effects 37
2.2 Interrelated L1 and L2 Effects 38
3 Methods 39
4 Results 41
4.1 Model Comparison 41
4.2 Control Variables 44
4.3 The L2 Distance Effect 44
5 Discussion and Conclusion 49
References 51
4 Autocorrelated Errors in Experimental Data in the Language Sciences: Some Solutions Offered by Generalized Additive Mixed Models 54
1 Introduction 54
2 Generalized Additive Mixed Models 55
3 Time Series in a Word Naming Task 56
4 Pitch Contours as Time Series 61
5 Time Series in EEG Registration 68
6 Concluding Remarks 73
References 74
5 Border Effects Among Catalan Dialects 75
1 Introduction 76
1.1 Border Effects 77
1.2 Combining Dialectometry and Social Dialectology 77
1.3 Hypotheses 78
2 Material 79
2.1 Pronunciation Data 79
2.2 Sociolinguistic Data 80
3 Methods 81
3.1 Obtaining Pronunciation Distances 81
3.2 Mixed-Effects Regression Modeling 83
3.3 Generalized Additive Mixed-Effects Regression Modeling 85
3.3.1 Social and Lexical Variables 86
4 Results 88
4.1 Demographic Predictors 89
4.2 Predictors Specific to Lexical Identity 92
4.3 Comparison to Individual Linguistic Variables 92
5 Discussion and Conclusions 97
References 99
6 Evaluating Logistic Mixed-Effects Models of Corpus-Linguistic Data in Light of Lexical Diffusion 102
1 Mixed-Effects Models in Corpus Linguistics 102
2 The Challenges of Corpus Data Given Lexical Diffusion 104
3 Purposes of Model Evaluation 106
4 The Problem 107
5 Simulations 108
5.1 Documenting the Problem 108
5.2 The Solution: Using Mixed-Effects Models to Derive Coefficients of the Evaluated Fixed-Effects Models When the Sample is Unbalanced 112
6 Limitations 115
7 Conclusion 115
References 116
7 (Non)metonymic Expressions for government in Chinese: A Mixed-Effects Logistic Regression Analysis 120
1 Introduction 120
2 Methodology 122
2.1 Data Collection 122
2.1.1 Corpus Design 122
2.1.2 Potential Expressions for government and Data Retrieval 123
2.1.3 Meaning Identification in Contexts 124
2.2 The Variables 129
2.2.1 The Response Variable Meto 129
2.2.2 The Predictors 129
2.2.3 Summary of the Variables 132
2.3 The Mixed-Effects Logistic Regression Model 133
2.3.1 The Random Effect: Verb 133
2.3.2 Model Selection and Diagnostics 134
2.3.3 The Regression Output 134
3 The General Regression Model for government 136
3.1 General Impact of the Predictors 136
3.2 Specific Influence of Fixed Effects 137
3.2.1 The Variable Con_gp and the Variable Style 137
3.2.2 The Variable Topic 139
3.2.3 The Variable Syn 139
3.2.4 The Variable Locus 140
3.2.5 The Variable LangVar 140
3.3 The Random-Effect Variable of Verbs 140
4 The Separate Regression Model for Mainland Chinese government 141
4.1 The Separate Mixed-Effects Model 142
4.2 The Lectal Variation Between Mainland and Taiwan Chinese 143
5 Summary 145
References 146