Book contents
- Frontmatter
- Contents
- Preface
- Notes on codes and abbreviations
- 1 Introduction
- 2 Data collection
- 3 The sociolinguistic interview
- 4 Data, data and more data
- 5 The linguistic variable
- 6 Formulating hypotheses/operationalising claims
- 7 The variable rule program: theory and practice
- 8 The how-to's of a variationist analysis
- 9 Distributional analysis
- 10 Multivariate analysis
- 11 Interpreting your results
- 12 Finding the story
- Glossary of terms
- References
- Index
9 - Distributional analysis
Published online by Cambridge University Press: 05 June 2012
- Frontmatter
- Contents
- Preface
- Notes on codes and abbreviations
- 1 Introduction
- 2 Data collection
- 3 The sociolinguistic interview
- 4 Data, data and more data
- 5 The linguistic variable
- 6 Formulating hypotheses/operationalising claims
- 7 The variable rule program: theory and practice
- 8 The how-to's of a variationist analysis
- 9 Distributional analysis
- 10 Multivariate analysis
- 11 Interpreting your results
- 12 Finding the story
- Glossary of terms
- References
- Index
Summary
How do you do a distributional analysis? Cross-tabs?
This chapter will cover how to conduct a factor by factor analysis.
It will also demonstrate how preliminary distributional analyses can pinpoint difficulties in research design and/or data anomalies.
It will focus on techniques for resolving data, computational and linguistic problems.
Now that you have some basic understanding of how the variable rule program works, let us now turn to the step-by-step procedures involved in performing an analysis. I will begin with distributional analysis.
FUNDAMENTALS
All too often when students first set out to do a distributional analysis, they do it the wrong way round. In order to do it right, distinguish between the roles of the dependent variable and the independent (explanatory) factors. Recall that in every variation analysis the focus is the tendency for the dependent variable to occur in a series of cross-cutting independent factors: ‘The essence of the analysis is an assessment of how the choice process is influenced by the different factors whose specific combinations define these contexts’ (Sankoff 1988c: 985).
THE WRONG WAY TO DO DISTRIBUTIONAL ANALYSIS
Many students make the mistake of reporting how the variants are distributed across the explanatory factors. Consider the results file for variable (t,d) in (1).
- Type
- Chapter
- Information
- Analysing Sociolinguistic Variation , pp. 191 - 216Publisher: Cambridge University PressPrint publication year: 2006