Description: a theory should bring order into the complexity of behaviours that have been observed and measured. It should help simplify, identify and clarify important issues.
Explanation: A theory should help in understanding the “why” of behaviour. Does the theory explain how and why individual differences in commonly observed instances of behaviour occur?
Value – Should stimulate interest in a research area. Should lead to beneficial changes – does it provide a qualitative leap in knowledge in a particular area? Does it take a new approach to solving problems?
Testable Concepts
Can the concepts in the theory be operationalised so that they can be tested – can the concept be defined precisely enough to enable it to be reliably measured?
Reliability
free from unsystematic error – the degree to which an assessment tool can produce stable and consistent results.
Some things we can consider include:
Test-retest reliability: Does the test give similar results at different times?
Parallel forms: Do equivalent test forms yield the same answer?
Split-half reliability: Cronbach's alpha: tests all possible halves and calculates the average correlation between halves