Questions: Reliability in Psychological Measurement

Question 1

A bathroom scale consistently reads 5 kg heavier than your true weight, every single time. How should this scale be characterized?

• A. Neither reliable nor valid
• B. Both reliable and valid
• C. Reliable but not valid
• D. Valid but not reliable

Answer: (C)

Reliability is about consistency — the scale produces the same result every time, so it is reliable. Validity is about accuracy — the scale does not measure your true weight, so it is not valid. This example illustrates that reliability is necessary but not sufficient for validity.

Question 2

A psychological measure with high reliability is necessarily also high in validity.

• T. True
• F. False

Answer: False

Reliability puts a ceiling on validity — an unreliable measure cannot be valid — but a reliable measure can still measure the wrong thing entirely. A scale reliably measuring wrist circumference is not a valid measure of intelligence, no matter how consistent its readings.

Question 3

What is internal consistency, and why is it important for multi-item psychological scales?

Model answer: Internal consistency is the degree to which items on a scale correlate with one another — indicating they are measuring the same underlying construct. It is typically assessed with Cronbach's alpha. It matters because if scale items are not intercorrelated, the total score is an incoherent mixture of different things rather than a consistent measure of one construct.

A depression scale should have items that all tap into depression. If some items measure anxiety and others measure fatigue without any shared variance, the composite score loses interpretability. Cronbach's alpha captures this by measuring the average inter-item correlation, adjusted for scale length.