 To do research, we need some data. The data comes from what we know as measurement. Singleton and Straits define measurement as a process of assigning numbers, and we could also say that it's a process of assigning numbers according to a known rule, to be more precise, and measurement can have different properties. The idea of measurement is that we typically start with some kind of theoretical concepts and then we need to come up with empirical measures of those theoretical concepts, and then we do calculations using those empirical measures. How it actually works, let's take a look. I'm going to use this diagram from Mikko Ketokivis' book, and it's based on Bach-Karach's article in Academy for Management Review. To analyze this diagram or use this diagram, we need a claim. And our claim or our proposition is that naming a woman as a CEO causes profitability to increase. And this is a proposition, it's a claim on the level of theory, and it contains two theoretical concepts. We have the first concept is CO-gender, the second concept is profitability. And this proposition here is basically a causal claim that female CEO causes profitability to go up. How do we then actually test this? So the idea is that if we have a proposition, then we need to have a way of assigning numbers to represent profitability and represent CO-gender. Then we make a hypothesis that these representations or empirical concepts are associated because we can't really observe causality. We can only observe associations. Then we collect some data and we test for statistical association. If we find a statistical association, then we conclude that we could not reject the hypothesis, therefore we found some evidence for the purposes. So how do we actually measure CO-gender? We could have every CEO to go through a medical examination and the doctor would determine their gender, or we could send them surveys, ask them to report their gender, but that's not practical. So how do we go in practice of determining which CEOs are men and which are women? One easy way would be looking at their names. So the names of CEOs that is public information and then we can check whether the name is a man's name or a woman's name and assign the gender variable according to that rule. So measurement is about assigning numbers according to the rule. Whether that the rule is entirely reliable and entirely valid is probably not because some names can be used for both genders, so there's a bit of unreliability, there's randomness in how a person evaluates them and also if there are names that are from different countries, they might be difficult for us to evaluate. So we get some data about specific companies, so that's the measurement process, and then we evaluate. How do we evaluate profitability? We need to define what is our measure and in this case our measure is return on assets. How do we claim that this is a valid measure of profitability? We can claim validity because ROA is actually something that investors look at when they look at profitability differences between companies. So it is valid in the sense that that is what people actually use. We can also argue validity of ROA based on understanding that profitability or performance of a company is related to how much money or how much profits that makes. And in ROA the returns is the profits and then assets simply scales those profits to be comparable across companies of different sizes. So we can do two different arguments, we can say that this is what people actually use for profitability or we can say that ROA can be derived from profitability and therefore it's a valid measure. Then we collect some data for companies and specific firms, we calculate some kind of sources. In this diagram, reliability is here, it's a very low level thing. Would we get the same data if we collected ROAs and Z-Ogenders again the next day? Probably yes, so this is probably highly reliable. Reliability also concerns the rating here. So if a person rates some companies as men-led or women-led based on name, would that be perfectly reliable? Maybe not. Maybe 99-95% reliable but certainly not 100% reliable because some names are gender ambiguous. And then validity concerns whether we can claim that actually name is valid information of gender. We can say that that is because of the tradition of naming boys in one way, girls in another way and then whether ROA is a valid measure of profitability which we can claim based on the definition on ROA or based on the fact that ROA is actually used as a measure of profitability by, for example, investors. So this is in a nutshell the idea of measurement. We have concepts and then we derive empirical representations for those concepts. We collect some data, we analyze association. If association exists, we conclude that we got some support for the theoretical purposes.