WEBVTT WEBVTT Kind: captions Language: en 00:00:00.030 --> 00:00:05.460 Measurement is an important part of social science research. If we take a look at this research 00:00:05.460 --> 00:00:11.700 process diagram from single demonstrates we can see that the two most important decisions after 00:00:11.700 --> 00:00:18.420 you have decided on your research question is what do you sample, what are the units of analysis and 00:00:18.420 --> 00:00:25.590 what do you measure which means are the variables that you study from those units of analysis. 00:00:25.590 --> 00:00:31.320 After that you make your data collection and you do your data analysis and report the results. 00:00:31.320 --> 00:00:39.330 The important part is that the quality of your study is mostly determined by what do 00:00:39.330 --> 00:00:43.680 you sample and what do you measure from your sample. So when you have your data 00:00:43.680 --> 00:00:49.560 then the upper limit of the quality of the study is basically determined 00:00:49.560 --> 00:00:54.930 If your measurement doesn't work or if your sample is somehow flawed then 00:00:54.930 --> 00:00:59.160 no matter how complicated or how sophisticated analysis you apply 00:00:59.160 --> 00:01:03.900 to those poor data your research output will not be very high. 00:01:03.900 --> 00:01:10.020 The idea of measurement is that we want to assign numbers to some quantities that 00:01:10.020 --> 00:01:16.350 we study. For example some things that we could study are heights of people, 00:01:16.350 --> 00:01:22.170 temperature outside, intelligence of a person, innovativeness of company. 00:01:22.170 --> 00:01:27.120 The idea of all these quantities is that they are variables. The idea 00:01:27.120 --> 00:01:31.530 that something is a variable means that it varies. Some people are taller than 00:01:31.530 --> 00:01:36.960 others. Sometimes it's colder outside. Sometimes it's warmer outside. Sometimes 00:01:36.960 --> 00:01:40.710 some people are smarter than others. Some companies are more innovative than others. 00:01:40.710 --> 00:01:45.870 So the idea is that there's some kind of variation in the objects 00:01:45.870 --> 00:01:49.830 or the units that you study and the idea of measurement is that 00:01:49.830 --> 00:01:56.130 you want to assign some numbers to that variation to quantify that variation. 00:01:56.130 --> 00:02:03.360 There are three key questions when you do measurement. The first question is where do 00:02:03.360 --> 00:02:11.220 you get the numbers? So how do we assign the height of a person? We can't quantify it. So 00:02:11.220 --> 00:02:17.520 for height that's obvious we use a measurement tape for example. For temperature you use a 00:02:17.520 --> 00:02:22.530 thermometer but there are different kinds of thermometers that you can apply also. 00:02:22.530 --> 00:02:29.160 But how do you quantify things that are not physical quantities like innovativeness or 00:02:29.160 --> 00:02:34.380 intelligence? That's a less straightforward to do and there are different ways of doing it. 00:02:34.380 --> 00:02:39.990 The next question is what does the number tell you? So if you say that a company's 00:02:39.990 --> 00:02:45.270 innovativeness is 5. Is it a lot or a little? What does it actually mean? 00:02:45.270 --> 00:02:50.520 So we're talking about the meaning of the number and the interpretation of the number. 00:02:50.520 --> 00:02:55.920 Finally how do we justify the way we assign the numbers? So we of course 00:02:55.920 --> 00:03:01.140 are besides just getting the numbers we have to convince our readers and ourselves 00:03:01.140 --> 00:03:05.850 that our numbers are actually valid for the purpose that we're using them for. 00:03:05.850 --> 00:03:12.720 There are a couple of different like higher-level ways of getting the numbers. 00:03:12.720 --> 00:03:19.020 Let's look at the research designs by Singleton and Straits. They present four research designs. 00:03:19.020 --> 00:03:22.770 The first is a laboratory experiment. The idea of 00:03:22.770 --> 00:03:28.080 laboratory experiment is that you don't actually measure the key variable that 00:03:28.080 --> 00:03:34.020 you're studying instead you manipulate it. So laboratory experiments and experimental 00:03:34.020 --> 00:03:38.250 studies are more about manipulation of things than measurement of things. 00:03:38.250 --> 00:03:42.150 The remaining three are about measurement and they 00:03:42.150 --> 00:03:47.070 are different approaches of measurement and to some extent sampling as well. 00:03:47.070 --> 00:03:52.410 The idea of a survey is that you measure things by asking people. 00:03:52.410 --> 00:03:58.140 So the subjects are provided the numbers. If we study people, their intelligence, 00:03:58.140 --> 00:04:03.780 then we ask them whether they're smart or not. And if we study companies we 00:04:03.780 --> 00:04:07.020 ask people in those companies whether the companies are innovative or not. 00:04:07.020 --> 00:04:12.990 We can do it also indirectly by asking whether the companies have been successful in producing 00:04:12.990 --> 00:04:17.820 new products and new services and then we have the second category field research. 00:04:17.820 --> 00:04:22.080 The idea of field research is that we don't ask the subjects instead we rate 00:04:22.080 --> 00:04:29.370 or our research assistant rates or evaluates the subjects and records 00:04:29.370 --> 00:04:33.270 what happens based on observation and that gives us the numbers. 00:04:33.270 --> 00:04:39.210 Finally we can use numbers collected by others so that's the archival records. 00:04:39.210 --> 00:04:46.290 So that's basically how we get the numbers. The actual practicalities of how they do that 00:04:46.290 --> 00:04:50.880 is something that I'll address in other video but those are the three main main 00:04:50.880 --> 00:04:56.850 ways of getting the numbers. Ask the people rate yourself. Use data collected by somebody else. 00:04:56.850 --> 00:05:05.640 The next question is what the numbers tell us and how do we justify the numbers? To 00:05:05.640 --> 00:05:10.620 answer those questions we need to understand a little bit about measurement theory which 00:05:10.620 --> 00:05:15.570 relates to how the data and the thing being measured are related. 00:05:15.570 --> 00:05:20.700 To understand measurement theory we need to understand the concepts of a latent 00:05:20.700 --> 00:05:26.070 variable. The idea of a latent variable is that an observed variable is that we 00:05:26.070 --> 00:05:32.100 have two types of variables. The observed variables are variables for which we have 00:05:32.100 --> 00:05:38.250 case values. So we have a specific number for each individual in our sample. We have 00:05:38.250 --> 00:05:42.600 a specific number for innovativeness of the first company's the second company and so on. 00:05:42.600 --> 00:05:49.530 These are in moral path diagrams. These are presented by these squares. Sometimes 00:05:49.530 --> 00:05:54.930 measurement measured variables are called indicators or manifest variables which 00:05:54.930 --> 00:05:59.430 highlight that their purpose of these measured variables is oftentimes to 00:05:59.430 --> 00:06:03.300 quantify some unmeasurable or unobservable thing. 00:06:03.300 --> 00:06:10.230 Latent variable is another kind of variable. The idea of latent variable is simply that it 00:06:10.230 --> 00:06:14.520 is a variable for which we don't have the case value. So we know that there is some 00:06:14.520 --> 00:06:20.850 variation between companies or between people but we cannot specifically assign 00:06:20.850 --> 00:06:24.450 numbers to any companies. We just know that there's some variation on 00:06:24.450 --> 00:06:29.190 some attribute or some variable but we cannot assign the exact numbers. 00:06:29.190 --> 00:06:34.410 We can estimate these numbers and we can estimate correlations within latent 00:06:34.410 --> 00:06:39.870 variables so we can't say what the specific values are. So the difference between latent 00:06:39.870 --> 00:06:46.140 variable and observed variable are important when we talk about measurement theory and when 00:06:46.140 --> 00:06:53.970 we talk about models that allow us to test or use or operationalize our measurement theory. 00:06:55.050 --> 00:06:58.080 Then we need to understand a couple of other terms as well. We have to 00:06:58.080 --> 00:07:01.230 understand the difference between concept construct and measure. 00:07:01.230 --> 00:07:10.710 The idea of a concept is that it's an abstract label for things that we study. And concepts have 00:07:10.710 --> 00:07:17.820 a reference and often the meaning as well. The idea of a reference is that for example if we have 00:07:17.820 --> 00:07:25.800 a concept of rock that refers to certain objects that we call rocks. The idea of meaning is that 00:07:25.800 --> 00:07:31.680 the concept has some kind of meaning as well. So if if we say that we have a rock then people know 00:07:31.680 --> 00:07:38.940 that well we have something that is a naturally occurring. It is a hard object. It's probably size 00:07:38.940 --> 00:07:45.240 of a couple of fists or so. That's it instead of a boulder which is a larger one and so on. 00:07:45.240 --> 00:07:49.650 So the term has some kind of attributes or the concept has 00:07:49.650 --> 00:07:54.630 some kind of attributes that are attached to it that give it meaning. Then a concept 00:07:54.630 --> 00:07:59.190 can also have a definition and the idea of a definition is that we have agreed 00:07:59.190 --> 00:08:06.000 on a specific written definition of what exactly the concept means. 00:08:06.000 --> 00:08:12.780 When you read papers that develop theory or introduce new constructs then they quite 00:08:12.780 --> 00:08:18.420 often define the construct explicitly. I'll get your constructs in a moment. 00:08:18.420 --> 00:08:24.960 Examples of concepts are persons for example and rocks and many other things. 00:08:24.960 --> 00:08:29.940 So concepts are like abstractions of things that we can observe and study. 00:08:29.940 --> 00:08:38.100 Construct is a special kind of a concept. It is a conceptual variable. So the idea is that 00:08:38.100 --> 00:08:45.510 it can vary so people or organizations can have different decrease of the cone structure. You 00:08:45.510 --> 00:08:51.330 can have a different decrease of innovativeness different amounts of intelligence and so on. 00:08:51.330 --> 00:08:56.790 So whereas in these concepts they can refer generally to just about anything, 00:08:56.790 --> 00:09:03.480 construct is something that is typically quantifiable. And the reason that these are 00:09:03.480 --> 00:09:08.610 because these are quantifiable we can study constructs using quantitative techniques. 00:09:08.610 --> 00:09:17.760 Constructs are also latent in the meaning that we cannot assign explicit correct values. We 00:09:17.760 --> 00:09:22.950 can only observe constructs indirectly. Constructs also can have dimensions for 00:09:22.950 --> 00:09:28.230 example we could have a construct of a person's size with the two main dimensions of height and 00:09:28.230 --> 00:09:35.130 weight of the person. Then some examples are intelligence. It could have some dimensions. 00:09:35.130 --> 00:09:41.970 Innovativeness it could have dimensions. So for example how well you are doing in product 00:09:41.970 --> 00:09:46.290 innovation and how well you are doing a service or process innovation and so on. 00:09:46.290 --> 00:09:53.760 Then measure is the third kind of on variable or a thing that you need to understand. Measure is an 00:09:53.760 --> 00:10:01.740 observable variable that quantifies one dimension of construct. If you have multiple dimensions in a 00:10:01.740 --> 00:10:07.920 construct then you need one at least one measure for each. So it doesn't make any sense to try to 00:10:07.920 --> 00:10:12.630 quantify person size using one number. You need at least two numbers: the height and weight. 00:10:12.630 --> 00:10:18.510 Examples include IQ test scores so that a measure for intelligence and 00:10:18.510 --> 00:10:23.400 reading on mercury column thermometer which is a measure of our temperature. 00:10:23.400 --> 00:10:31.380 How do these construction measures then relate? There are two main approaches. 00:10:31.380 --> 00:10:38.970 One is a nominalism. The idea is that in nominalism you basically reject the 00:10:38.970 --> 00:10:45.360 existence of constructs independently of measurement. An extreme version of 00:10:45.360 --> 00:10:52.890 nominalism is operationalism which says that the construct is simply whatever the 00:10:52.890 --> 00:10:57.630 measurement process produces. So the construct is defined by the measure. 00:10:57.630 --> 00:11:05.880 And then realism assumes that the constructs exist independently of measurement and the purpose of 00:11:05.880 --> 00:11:13.020 measurement is to discover the true values of the construct. Most social science research follows 00:11:13.020 --> 00:11:18.840 the realist approach. So the idea is that there exists something called innovativeness 00:11:18.840 --> 00:11:24.510 independently of our measurement. Some we can say that some companies are more innovative 00:11:24.510 --> 00:11:30.660 than others without measuring those. So that kind of statements make sense if we assume 00:11:30.660 --> 00:11:37.770 that innovativeness or intelligence exist independently of our measurement attempts. 00:11:37.770 --> 00:11:47.190 Then how we actually apply these concepts in practices is that we use the measures as 00:11:47.190 --> 00:11:54.330 proxies for the constructs. We cannot really observe the constructs directly so the next 00:11:54.330 --> 00:12:01.590 best best thing is that we build some kind of statistical presentation based on our data. 00:12:01.590 --> 00:12:08.700 And for example or we can just use a number as such, we can take a sum of multiple numbers or 00:12:08.700 --> 00:12:16.230 we can build a latent variable model and then use the latent variable as a proxy for the construct. 00:12:16.230 --> 00:12:23.010 So we use these empirical representations constructed based on our data as proxies 00:12:23.010 --> 00:12:30.570 for the constructs assuming that the implicit representation is a perfect 00:12:30.570 --> 00:12:35.340 representation of the construct. That is of course something that is a hardly ever 00:12:35.340 --> 00:12:39.600 exactly correct but we have to justify that it is a good enough approximation. 00:12:39.600 --> 00:12:44.610 So that's the idea of a proxy. Instead of using the construct when we study 00:12:44.610 --> 00:12:48.840 something we use the measure as a stand-in for the construct. 00:12:48.840 --> 00:12:57.450 Summary of these key concepts. We have the constructs. Construct is a concept so it's 00:12:57.450 --> 00:13:02.610 a variable that exists in principle. It can have some definition. Almost 00:13:03.120 --> 00:13:10.080 always has a definition. We can say that some companies or some individuals are 00:13:10.080 --> 00:13:14.610 higher on the construct than others and we cannot observe it directly. 00:13:14.610 --> 00:13:22.500 Then observable variables are specific numbers for each subject or a case that we have collected 00:13:22.500 --> 00:13:30.360 somehow. The idea is that these measures are - if we take the realist perspective - the 00:13:30.360 --> 00:13:37.170 idea is that the variation in these measures is caused by the variation in the construct. 00:13:37.170 --> 00:13:44.910 For example our people's IQ score differs because their intelligence differs. So the 00:13:44.910 --> 00:13:50.130 reason why there's variation in the data is that there's variation in the construct. Thermometer 00:13:50.130 --> 00:13:56.550 changes its value because the temperature outside is different from one day to another. 00:13:56.550 --> 00:14:03.090 So that's the idea of realist perspective to measurement which I will be using in these videos.