In this video, I cover properties, operational definitions, and construct validity. Understanding these concepts is fundamental for determining which conclusions can be drawn from a study and should always be kept in mind when considering interpretations of research.
Don’t forget to subscribe to the channel to see future videos! Have questions or topics you’d like to see covered in a future video? Let me know by commenting or sending me an email!
Need more explanation? Check out my full psychology guide: Master Introductory Psychology: http://amzn.to/2eTqm5s
Viewers in China can find this video here: http://v.youku.com/v_show/id_XMTgzOTA4NzM5Ng
Video transcript:
Hi, I’m Michael Corayer and this is Psych Exam Review. In this video I’d like to talk about operation definitions and construct validity. I’m going to start with a property.
So what is a property? A property is a general sort of concept that we want to investigate. So researchers might want to know about the property of happiness or they might want to know about depression or intelligence. And an important thing about properties is that we can’t directly detect them so I don’t have a happiness detector that’s going to tell me exactly how happy you are.
Instead I need to have a way of defining it and this is my operational definition. So an operation definition is the particular way that I’m going to measure the property. It’s important to remember that it’s not actually the property, it’s just a representation of the property.
It gives us a way to talk about it, so for instance if I wanted to know about the length of a desk you could tell me that the desk is 80 centimeters. And now this is not saying that 80 centimeters on a tape measure is the desk but it’s a useful representation to help you determine whether the desk is going to fit in a particular space and there’s other operational definitions that you could use to talk about the length of the desk. You could tell me that it’s 0.000497097 miles and that might be just as accurate as telling me it’s 80 cm but in this particular situation it’s not that useful. So different operational definitions are going to be more appropriate in certain situations.
In contrast, if I wanted to know the distance between Austin and Houston, I’d probably prefer that you tell me in miles, because I’m not used to thinking about how many centimeters per hour my car can travel. So if I want to figure out how long this trip is gonna take, centimeters isn’t very useful.
Now we have experience using certain operational definitions like for length we’re used to using cm or inches or miles so we have a pretty good idea what that means but when it comes to psychological variables it can be little bit more complicated.
So let’s say we want to know about happiness. How am I going to define happiness? I can’t detect it directly so I need to come up with an operational definition, something that I can actually measure and that’s going to represent it. So you might say “well, ok I could ask people on a survey I could say, how happy are you from -5 to +5”, that would give me a measurement.
But it might be kinda hard to determine what that means. What is +3? That’s not as comfortable to me as using cm or miles. I don’t really know what +3 happiness means. Ok, well I might measure smiling behavior. People smile when they’re happy, people are happy they tend to smile, so I’m going to say that, I’m going to count how much people smile and use that to determine how happy they are.
That might be a way of defining happiness. Or I might say smiling is too easy to fake, I want to use laughing so I’m going to show people a video, count how many times they laugh, use that to determine how happy the video made them.
You might say, well there might be some brain activity associated with happiness. I’m going to measure people’s brain activity in a scanner and show this certain pattern of activity and use that to represent happiness.
What we’re hoping for with any of these operational definitions, we want to have construct validity and what construct validity refers to is the idea that we have the property and we want to see a clear relationship between the property and the operational definition we chose.
So I’m going to look at happiness and smiling, is there a clear relationship between these two? And we hope that there is. We hope that the property is actually related to the thing that we’re measuring. So let’s use that example, let’s say OK I want to know about happiness and I’m going to measure smiling behavior and we’re going to ask, is there a clear relationship between these two things?
Ok, people smile when they’re happy, if people are unhappy they tend not to smile so much, so OK this seems like it has construct validity.
There seems to be a clear relationship between this property and this definition. It’s not perfect. Certainly people can be happy and not smile and people can be smiling even though they’re not actually happy. So it’s not perfect but we can say it has construct validity, it seems to have a clear relationship between these two things.
In contrast, I might say I wanna know about happiness and let’s see, I was really happy when I got a car so I think cars make people happy, so I’m going to count how many cars people have and I’m going to use that to determine how happy they are. If you have more cars then you’re happier, if you have fewer cars, not so happy.
In this case you might look at this relationship and say “I’m not so convinced here. I don’t think there’s a clear relationship between happiness and the number of cars that somebody owns”.
There might be some relationship but it’s not a very clear one. And I can certainly imageine lots of people who don’t have any cars who bike everywhere and they seem perfectly happy and I can imagine somebody who has dozens of cars but isn’t actually happy. So in this case, I’m going to say I’m not convinced, I don’t think we have construct validity here. Our operational definition is not a good representation of our property.
Now this is where we see that operational definitions and construct validity are extremely important because if this relationship isn’t clear then my conclusions don’t really matter. If I told you my study was on happiness and I found some conclusion about happiness and then you looked at my paper and said “wait a minute, you defined happiness as the number of cars that people had?” you would immediately disregard my results, they wouldn’t matter because you’d say “they’re not really about happiness”.
So this is why it’s so important, a faulty operational definition with low construct validity causes us to basically disregard any of the results from that study. This is why this should be the first thing that you look at whenever you see a study or you see some results. You say, “wait a minute, what’s the property and how are they measuring it?”. If that relationship isn’t clear then I don’t really care what the results were.
I hope you found this helpful, if so please like the video and subscribe to the channel for more. Thanks for watching!
