[MUSIC] So today is the talk that I usually get asked to give, which is on Grit. And here is the Grit Scale, which I gave out in hard copy last time. One of the things I want to tell you about is that I spent a long time validating the scale, right? And the one syllable synonym for validity is >> True. >> True. Like is it truly measuring passion or perseverance? So I accumulated evidence, it was very, very boring, it's like extremely is there convergent validity? Does it converge with other measures of persistence? If somebody else has a scale, it's not quite what I want, but it does get it perseverance or persistence. Does it correlate with my scale? Yes, what about a discriminative validity? Is it truly different from IQ? Is it truly different from leadership? So I had to spend a lot of time showing that it's discriminate, it's different from these other things that it's not supposed to be like. And then predictability, does it actually predict the things it's supposed to predict? Does it predict staying at West Point through the first summer of training? Does it predict winning harder things like the spelling bee? That was another few years of that. So, convergent validity, discriminant validity, predictive validity, three ways of thinking about is it truly measuring what it's supposed to be measuring? But there's a fourth kind of validity that you're going to learn about at some point and it's called consequential validity. So what is consequential validity? So like so many times in your future, you're going to see something that's kind of familiar, like validity error, I remember validity, because I took Map 601, but I don't remember what consequential validity is. And you will do what all my graduate students do, which is you'll Google it, right. Which is why it says Google it behind my desk This- >> [LAUGH] >> When I come in, and they're like, Google it. But when you Google consequential validity, what you'll find out is that it means is it true? Is it useful for the consequences that you're using it for? So for example, when NFL coaches ask me whether they can use the Grit Scale instead of the NFL Combine I say no. >> [LAUGH] >> Because it's pretty fakable.. I am a hard worker. I finish whatever I begin. Fine, you don't need a manual to figure out how to get a high score on it. So that's a failure of consequential validity, don't use the scale in a high stakes situation where there's an incentive to fit. We talked about cross-cultural comparisons, I could, for example, administer the Grit Scale in 20 countries and then I could try to publish an article on who's the grittiest country, right? I mean, it sounds really sexy, right? Who is the grittiest country, is it Finland, is it the United States, is it Japan? But I wouldn't do that because I think, again, it's not valid for that purpose because when you read the scale in one country and you're thinking about what it means to be a hard worker, you have a completely different reference than in another country. So, generally, psychologists don't believe that self report questionnaires have a lot of consequential validity for cross-cultural research. So, that's also on the do not do list. But there are consequences or uses that we do believe in, like I think the scale is useful within a culture when there's no incentive for faking to do research on what correlates with individual differences in Grit. That's actually what I've been using it for. And also, I believe that it's consequentially valid for just reading it and learning and I think a lot of the things that you have been doing as a student. You're going to go off and have other people take the pannus or the life satisfaction scale. And it's not like you're collecting their data and showing them how to do a correlation or doing a correlation yourself. It's because there's something illuminating, I think, about answering questions like this. Like so I believe that there's consequential validity. I believe that there's consequential validity for reflection, but I don't believe in almost any other consequence, including high stakes settings where faking is incredibly easy and incentivized, for lack of a better word.