Chapter 8: The Nature of Quantitative Research

The nature of Quantative Research

Reliability

Concerns the consistency of measurements and includes:

Stability

is assessed through the test-retest method, where a measure is administered twice to the same group, and a high correlation between the two administrations indicates stability. However, this method has limitations due to potential response influence and external factors.

Internal reliability

evaluates whether the indicators within a concept measure are consistent with each other. The split-half method is one way to assess this, aiming to show that respondents who score high on one set of indicators also score high on another set.

Inter-rater reliability

Is relevant when subjective judgment is involved, such as in content analysis or structured observation. It concerns the consistency of decisions made by different raters and is explored more in advanced contexts.

Validity

assesses if a measure truly captures the concept.

Face validity

which gauges whether the measure intuitively reflects the concept. Experts or

judges may evaluate this.

Concurrent validity

where the measure's results are compared with a known criterion that relates to the concept. For instance, assessing job satisfaction by comparing it to absenteeism.

Predictive validity

which uses future criteria rather than concurrent ones. Future levels of absenteeism might be used to validate a job satisfaction measure.

Convergent validity

involves comparing the measure with others that assess the same concept through different methods.

Discriminant validity

ensures that the measure is distinct from measures of other, related constructs.