Though reliability and validity are different from each other, they are still related.
In this lesson, we’ll look at the differences of and relationship between reliability and validity.
Imagine that you are a psychologist, and you want to study the relationship between stereotyped thinking and critical thinking. You believe that people who think in terms of stereotypes will have lower critical thinking skills than people who do not employ stereotypes in their thinking. But, how do you know whether a person thinks in terms of stereotypes or not? How do you know if they have high critical thinking skills?Psychological measurement is the process of assessing psychological traits.
These could be things like stereotyped thinking or critical thinking skills, or they could be traits, like anxiety, optimism or intelligence. When measuring anything, there are two major issues that need to be taken into account: reliability and validity. Let’s look closer at both of them and how they are related to each other.
You wake up one morning and step on the scale. It reads ‘117,’ which you think is too low. You step off the scale and back on again, and it reads ‘143.’ Okay, well, that’s better, but just to be sure, you step off and back on for a third time. This time, the scale reads ‘108.
‘I hate to break the bad news, but your scale is not reliable. Reliability is the extent to which a measurement tool gives consistent results. If your scale read ‘143’ every time you stepped on it, it would be reliable. But, notice that the scale doesn’t have to be right to be reliable; it just has to produce consistent results. For example, if your scale read ‘108’ every single time you stepped on it, but you really weigh 143, your scale is reliable even though it’s not correct.
Let’s say that for our study on stereotyped thinking, we come up with a survey that measures stereotyped thinking. If I take that survey one day and then take it on another day, my score should be about the same. If one day I take it and the survey says I think in stereotypes a lot, and the next day it says I never think in terms of stereotypes, we have a reliability issue!But, reliability isn’t the only thing that’s important in measurement.
Imagine that you step on your scale, and it consistently reads ‘102.’ You know that you don’t weigh 102 pounds, so what’s going on? Well, it turns out that the scale is measuring the IQ of the person standing on it, not his or her weight.The extent to which a measurement tool measures what it is supposed to measure is called validity. In the case of your bathroom scale, it has a low validity because it’s supposed to measure your weight, but it’s really measuring your IQ, something completely different.A measure can be reliable but not valid. For example, if our survey about stereotyped thinking had a high reliability, it would consistently give the same answer.
But, if it wasn’t measuring stereotyped thinking but instead measuring something else (say, IQ), it would have a low validity.
Because a tool can be reliable but not valid, many people mistakenly believe that they are unrelated. But, there is actually a relationship between reliability and validity. It’s just a complicated one. To simplify it, let’s imagine for a moment that we’re playing darts.
We have a target in front of us, and we’re trying to hit the center of the target. Imagine that the center of the target is the thing that we’re trying to measure, like stereotyped thinking.There are three possible outcomes:1. We are far away from the target, but all of our darts land close to each other.This is an example of when we might have high reliability but low validity.
We are being reliable with all our throws: all the darts are landing near each other. The problem is they aren’t landing near the target (the thing we want to measure). Because they aren’t close to the thing we’re measuring, we have low validity.2. Our darts are spread out across the whole board.
In this case, we are neither reliable nor valid. Each dart lands in a different part of the target, which means that we are not throwing with reliability. At the same time, if we hit the center, it’s just by chance, so we are not valid.3.
Our darts are clustered really close together near the center of the target.This is the ideal situation; we have a high reliability and high validity. All of our throws are landing close together, which shows that we are being reliable. In addition, they are landing near the center of the target, which represents the thing we want to measure.
Because we’re close to that, we have high validity.So, our three options are high reliability-low validity, low reliability-low validity or high reliability-high validity. Notice that we do not have an option for low reliability-high validity because in order to be valid, a measure has to be reliable.
Psychological measurement is the process of measuring psychological traits.
In measurement, two important concepts are reliability, or how consistent the outcome of the tool is, and validity, or how well the tool measures what it is supposed to measure. There are three possible relationships between reliability and validity: a measure can be reliable but not valid, neither reliable nor valid or both reliable and valid.
Following this video lesson, you will be able to:
- Define reliability and validity
- Explain the importance of reliability and validity to psychological measurement
- Identify the three relationships that can exist between reliability and validity