Sentence Repetition Testing
| Data Collection Tools | |
|---|---|
 | |
| Interviews | |
| Observation | |
| Questionnaires | |
| Recorded Text Testing | |
| Sentence Repetition Testing | |
| Word Lists | |
| Participatory Methods | |
| Matched-Guise |
Contents
Introduction
A Sentence Repetition Test (SRT) is a tool which langauge surveyors can use to assess community bilingualism. The tool consists of a series of increasingly more complex sentences. The language of the sentences is not the first language of the participant. As the participant hears each sentence, they are asked to repeat it. The surveyor notes the number of errors made in this repetition. The ability of the participant to repeat the sentences and the number of sentences they can repeat without error indicates how bilingual they are in the language of the sentences.
The test is based on the fact that, for known languages, our brain stores and processes languages in sequences or so-called chunks. A fluent language user has these chunks stored in long-term memory and will be able to reproduce and process them far faster than someone who is a learner. Someone who does not know the language at all may be able to repeat a certain amount, but the short-term memory load will soon be overwhelmed by longer and more complex chunks of language.
As with all assessment tools, findings based on SRTs should be triangulated with other findings from other tools. Data from SRTs are unlikely to be dependent enough for conclusive findings to be based on them alone.
Much of the information below is based on Carla Radloff's excellent overview of SRTs <ref>Radloff, Carla F. (1991). Sentence Repetition Testing for Studies of Community Bilingualism. Dallas: Summer Institute of Linguistics.</ref>
A Bit of History
In the 1980s, a vast sociolinguistic survey of northern Pakistan was carried out with Lok Virsa, the National Institute of Folk Heritage in Islamabad. While working on this survey, SRTs were developed in order to improve methodology and get a better grasp of how bilingual a community was.
Prior to this, second language proficiency testing in language assessment was largely based on tests which were not ideally suited to the survey context. As is still the case in formal second language learning environments, tests are often highly complex and run by highly trained personnel in carefully controlled environments. Attempts to use this in assessment fieldwork revealed a number of methodological problems, largely centring around how difficult it was to administer such a demanding test consistently. Consequently, accurately generalising results to the wider community was problematic.
Preparation
Is SRT the right tool for the job?
Before deciding to use the SRT in language assessement, it's important to make sure that the tool is relevant to the data your survey needs to collect. It's essential to determine that information about proficiency in a second-language is needed before using SRTs, particularly if the question is to what extent groups of a community might differ in their proficiency.
To what extent does it need to be used?
Secondly, you should find out how much testing needs to be done. Usually in language assessment, we are attempting to find out if the population is unable to function at a certain level of procificiency. If they cannot, it is likely that they will need materials in their own language. If however, they can, further testing will be necessary to see to what extent materials in the second language might be useful for them. This might involve some form of extensibility testing.
How do you select participants?
Sampling should be used to select participants who should also be people who have no impediment to speaking clearly i.e. all their teeth, aren't chewing something, etc.
It's important to realise that SRTs are not tests of inherent intelligibility. When testing to see if people can understand another language because of linguistic relatedness between their L1 and an L2, testing only a few speakers of the L1 will give you the information you need. This is because the variables lie not in the speakers themselves but in the languages themselves. It is thus assumed that among any group of L1 speakers, the language they speak will be relatively uniform and so testing only a small number of any of them will give the same results as if we tested all of them.
But SRTs are not used to test inherent intelligibility. They are used to test language proficiency... and proficiency varies not only from person to person but also from day to day for each language user. This means we need to sample well also to bear in mind that our results may be influenced by variables as idiosyncratic as time of day or whether the participant's baby kept them up for hours the night before!
<references />
