Reliability and Validity 

At 2:14 pm, the thermometer on the left "measured" the temperature at 51 degrees, whereas the one on the right "measured" the temperature at 54 degrees.
Can both be correct?


From this picture alone, can you tell whether the difference between the gauges is due to random error or to bias?


If we had only one gauge (only one "observer"), would we have known that we had a measurement problem?


If both gauges are accurate, should they both give us the same reading?


If two observers are objective, should they both give us the same measurement?


If both gauges gave us the same reading, must they both be right?


If two observers gave us the same measurement, must they both be right?


If one gauge was consistently (always) read 3 degrees lower than the other, the inter-observer correlation would be low.


If one rater was consistently (always) rating participants 3 points lower than the other, the inter-observer correlation would be low.


High inter-observer reliability means that your measure is free from bias.


If inter-observer reliability is low, then reliability will be low.



Back to Overview of this material

Back to Chapter 5 Menu

Back to Research Design Explained Home Page