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 because of cultural bias that can be present in the questions
asked. Let’s look at an example.
SCENARIO
To test the intelligence of university students, several colleagues devised a questionnaire to assess students' problem-solving ability. Two of the questions in the survey were:
Estimate the length of the Sydney Football Stadium.
Who wrote Hamlet?
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.
