 | |
|
| Describing, Clarifying and Presenting Data
2. Characteristic, Variable and Measurement > 2.1. The relationship between characteristic and variable
2.1.3. Selecting an appropriate variable to measure a characteristic
When you are thinking about variables, you need to ask the following questions:
- How is the characteristic defined by the variables?
- What purpose is being served?
- Is the measurement valid for the population?
If the characteristic is not directly observable, then it will be unlikely that one variable can completely represent the characteristic. Consider the use of IQ tests to measure intelligence (characteristic/ variable). IQ tests can be criticised because they only measure aspects of intelligence such as verbal or numerical intelligence and cultural bias can therefore be present in the questions asked. Let’s look at an example.
REMEMBER: You need to think very carefully about how variables in a study are constructed and how they are measured to decide if they are valid.
| |
|
This page last updated 8 October 2009 | 
