Summary

This article continues the methodological reflections on the variationist paradigm in the first part. While in Part I we tried to describe the most successful tool for gathering data – the interview - on this occasion we attempt to describe what could be the pattern for overall research. We will base this on the pattern supplied by C. Silva-Corvalán (1989).

Summary

1. Observation of the speech community and working hypothesis

2. Selection of informants
2.1. Stratification of the population
2.2. The sample
2.3. Access to the informants
2.4. The participants

3. Data gathering. Complementary techniques to the sociolinguistic interview
3.1. Anonymous observation
3.2 The subjective evaluation trial

4. Quantitative treatment
4.1. Data codification
4.2. Models

5. Bibliography

1. Observation of the speech community and working hypothesis

Variable phenomena are always observed within the framework of a given speech community. They may be seen in many different ways, from the simple confirmation of its occurrence in a communicative situation to discovery of information arising from prior research. However it occurs, when undertaking variationist research it is highly advisable – indeed, we feel that it is essential – that exploratory studies be carried out.

This research should be understood as stages prior to more in-depth investigation. Its results will therefore be partial, but will help us in the orientation of the work. The most important points usually refer to the establishment of the variable to be studied and the context of occurrence. In this way, the researcher already starts to have enough sufficiently relevant ideas to establish the working hypothesis.

By definition, the hypothetical-deductive method is based on provisional considerations that must be validated (or invalidated). These prior suppositions range from very general approaches to highly concrete expectations. An example of the former could be the hypothesis of variability, according to which individuals’ speech in a given community is characterised by the presence of certain variable linguistic traits, with a frequency of appearance depending on independent linguistic or extra-linguistic factors. It is these exploratory studies which enable more concrete hypotheses to be formulated regarding the factors that encourage or restrict the application of the resulting variable rule.

Other fields where exploratory studies provide very valuable information are the stratification of population and the design of the most suitable data gathering techniques. In short, if sure-footedness is desired on the tortuous path of variation we should carry out the prior reconnaissance necessary to travel down it in the correct direction.

2. Selection of informants

The correct choice of informants is of vital importance in methodological terms. It must be carried out according to the objective of the research. It is obvious that if we plan to carry out an analysis of sociolinguistic correlations, the choice of informants will have determining factors notably different to those in studies of phenomena characteristic of linguistic interaction in multilingual communities (change of code, grammatical convergence, etc.)

It is vital to determine whether investigation will at the outset be centred on linguistic factors or whether the initial hypothesis anticipates the participation of social and stylistic factors. While in the first case the sample may be homogenous in terms of age and level of education, in the second it will be necessary to establish the social and stylistic factors to be taken into consideration.