Difference between revisions of "Analysing SRT Data"
| (5 intermediate revisions by the same user not shown) | |||
| Line 6: | Line 6: | ||
==Initial Analysis== | ==Initial Analysis== | ||
| − | Each SRT has a total of 45 possible | + | Each SRT has a total of 45 possible points. During the [[Developing an SRT|development of the SRT]], you will have been able to create score ranges which define each level of proficiency that you are testing. Thus, you can take each participant's score and then see which proficiency level their score falls into. |
Once you have done this for all the participants, you will need to calculate the percentage of participants you have at each level of proficiency. Remember, you are not assessing individuals for their own sake. You are assessing the proficiency of the community and so you need to put all participants' results together so that collectively, they enable a picture of the entire community to emerge. | Once you have done this for all the participants, you will need to calculate the percentage of participants you have at each level of proficiency. Remember, you are not assessing individuals for their own sake. You are assessing the proficiency of the community and so you need to put all participants' results together so that collectively, they enable a picture of the entire community to emerge. | ||
In addition, because survey does not rely on only one tool in order to get a picture of the sociolinguistic reality of the people you work with, you will perhaps already have an idea of what the results should demonstrate. This impression can be used to triangulate the findings that the SRT reveals and/or to give you some idea as to why proficiency of the community in the test language is as the results imply. | In addition, because survey does not rely on only one tool in order to get a picture of the sociolinguistic reality of the people you work with, you will perhaps already have an idea of what the results should demonstrate. This impression can be used to triangulate the findings that the SRT reveals and/or to give you some idea as to why proficiency of the community in the test language is as the results imply. | ||
| + | |||
| + | ==Extrapolating Results== | ||
| + | The data you have collected represents only a fraction of the community. But if you have [[sampling|sampled]] well, it will be a representative sample and should have provided you with data that would have been similar if not identical if you had tested everyone. The process of taking a small set of data and using it to say something about the entire community is called extrapolation. If you understand extrapolation, you can see how sampling is so important. If we get the sampling even slightly wrong, we are likely to make even larger errors in our understanding of the larger population. | ||
When you carried out the test, you were able to collect demographic data on each participant. Thus, you can group your results so that you have some idea of, say, educated males and what their proficiency is. If you also collected information from each test site about what proportion of males are educated, you will be able to combine these figures to give an overall proficiency indication for all educated males. The exact demographic groups that you are interested in will vary from survey to survey but you can see how important it is to be aware in advance what you want to know about a community so that you can collect the right data and analyse it appropriately afterwards. | When you carried out the test, you were able to collect demographic data on each participant. Thus, you can group your results so that you have some idea of, say, educated males and what their proficiency is. If you also collected information from each test site about what proportion of males are educated, you will be able to combine these figures to give an overall proficiency indication for all educated males. The exact demographic groups that you are interested in will vary from survey to survey but you can see how important it is to be aware in advance what you want to know about a community so that you can collect the right data and analyse it appropriately afterwards. | ||
| Line 16: | Line 19: | ||
There are many different ways to extrapolate data. Some use statistical procedures and others do not. Whichever way you choose, you need to bear in mind what the purpose of your survey is so that your method provides the right analysis. | There are many different ways to extrapolate data. Some use statistical procedures and others do not. Whichever way you choose, you need to bear in mind what the purpose of your survey is so that your method provides the right analysis. | ||
| − | + | [[Category:Sentence_Repetition_Testing]] | |
| − | |||
Latest revision as of 17:24, 12 July 2011
| Data Collection Tools | |
|---|---|
 | |
| Interviews | |
| Observation | |
| Questionnaires | |
| Recorded Text Testing | |
| Sentence Repetition Testing | |
| Word Lists | |
| Participatory Methods | |
| Matched-Guise |
| Sentence Repetition Tests | |
|---|---|
| Developing an SRT | |
| Administering an SRT | |
| Analysing SRT Data |
Once you've returned from your survey or have carried out a batch of SRTs, you will be able to look at the results and calculate proficiencies for the participants you've tested.
Initial Analysis
Each SRT has a total of 45 possible points. During the development of the SRT, you will have been able to create score ranges which define each level of proficiency that you are testing. Thus, you can take each participant's score and then see which proficiency level their score falls into.
Once you have done this for all the participants, you will need to calculate the percentage of participants you have at each level of proficiency. Remember, you are not assessing individuals for their own sake. You are assessing the proficiency of the community and so you need to put all participants' results together so that collectively, they enable a picture of the entire community to emerge.
In addition, because survey does not rely on only one tool in order to get a picture of the sociolinguistic reality of the people you work with, you will perhaps already have an idea of what the results should demonstrate. This impression can be used to triangulate the findings that the SRT reveals and/or to give you some idea as to why proficiency of the community in the test language is as the results imply.
Extrapolating Results
The data you have collected represents only a fraction of the community. But if you have sampled well, it will be a representative sample and should have provided you with data that would have been similar if not identical if you had tested everyone. The process of taking a small set of data and using it to say something about the entire community is called extrapolation. If you understand extrapolation, you can see how sampling is so important. If we get the sampling even slightly wrong, we are likely to make even larger errors in our understanding of the larger population.
When you carried out the test, you were able to collect demographic data on each participant. Thus, you can group your results so that you have some idea of, say, educated males and what their proficiency is. If you also collected information from each test site about what proportion of males are educated, you will be able to combine these figures to give an overall proficiency indication for all educated males. The exact demographic groups that you are interested in will vary from survey to survey but you can see how important it is to be aware in advance what you want to know about a community so that you can collect the right data and analyse it appropriately afterwards.
There are many different ways to extrapolate data. Some use statistical procedures and others do not. Whichever way you choose, you need to bear in mind what the purpose of your survey is so that your method provides the right analysis.
