Language testing and methods of assessment
Standardised tests: psychometric-objective and linguistic-realistic
Methods of Assessment
16. What are standardised tests: psychometric objective and linguistic realistic?
What are standardised tests: psychometric objective and linguistic realistic?
An objective test is one that has been tried out on an appropriate sample of the population for whom it is intended. The test is then evaluated to establish whether it is both reliable and valid.
It is usually made up of individual items, each of which can be scored independently and which contributes to the candidate's overall performance. The items should be linguistically meaningful and appropriate to the purpose of the test.
Standardised scores
See Hatch & Farhady, Chapter 7, Standardised Scores.
Standardised scores are based on the concept of a normal distribution. In a normal distribution, scores are distributed symmetrically around the mean, producing the familiar bell-shaped curve. The mean, median and mode are the same.
In a normal distribution:
- approximately 68.26% of scores fall within one standard deviation of the mean;
- approximately 95% fall within two standard deviations;
- approximately 99% fall within three standard deviations.
A z-score shows how far a particular score is from the mean in terms of standard deviations:
By converting raw scores into z-scores, it is possible to obtain a common unit of measurement, even when the original scores were obtained from different tests or used different scales.
In an IQ test, the mean is conventionally set at 100 rather than at zero, as it is with a z-score. This provides a convenient scale for expressing scores above and below the mean.