Hi. This tutorial is all about variables. A variable is a measurable factor, characteristic, or attribute of an individual or a system. So here are just some examples of variables. We could have shoe size, first name, the amount of change in your pocket, gender, race, number of books in backpack, religious affiliation, age, temperature, zip code. These would all be considered variables. These are either factors or their characteristics or their attributes of something or a combination of things, that being a system.
When a cause and effect relationship is being explored between two variables, it is necessary to identify which is the explanatory variable and which is the response variable. So let's define both of those two terms, and then we'll look at some pairs of variables.
The explanatory variable is the variable that causes an effect, sometimes called the independent variable. So the explanatory variable is what explains or causes some effect. Now the response variable is the variable that reflects the effect, sometimes called the dependent variable. So independent goes with explanatory. Dependent goes with response.
So now what we're going to do is look at a couple different pairs of variables and try to figure out which is the explanatory variable and which is the response variable. So the first pair I have here is lifetime and weekly cigarette consumption. So it's widely known that cigarettes do have a detrimental effect on lifespan.
So cigarettes or cigarette consumption is going to cause a change or cause and effect in lifetime. So here, my weekly cigarette consumption is the explanatory variable. Lifetime is the response variable. So I'm going to use exp for explanatory and res for response. So again, cigarettes cause a change in lifetime or explain a change in lifetime.
Our next pair is daily high temperature and daily air conditioning energy consumption. So generally, as the temperature goes up, as it gets hotter and hotter, air conditioning consumption is going to go up. So what we're going to say here is that daily high temperature causes or explains a change in daily air conditioning energy consumption. So now we're going to call this one the explanatory variable. And we're going to call energy consumption due to air conditioning the response variable.
Our 3rd pair, we have size and calories of fast food sandwiches. So if we think about the relationship between those two variables, the size of the sandwich and the number of calories, generally, the larger the sandwich-- if we're looking at a double quarter pounder. It's a very large sandwich-- it's also going to be high in calories. So the size of the sandwich is going to explain a change in the calories. So here, size is explanatory. Calories is the response variable.
Our fourth pair here, airfare and distance to destination. Now this probably doesn't have a really strong relationship. But in general, the further you need to fly, the costlier it's going to be. So if you're flying from California to Australia, it has a very far distance, and that's going to be a very expensive plane ride.
So what I would say here is that the distance the destination explains a change in airfare. And certainly, there are going to be counter examples for each of these. But I would definitely say that airfare doesn't explain distance. Distance explains airfare.
And our last pair here is latitude and average January temperature of US cities. So latitude measures how far north or south you are on the globe. So we know that as the further you go north, the colder the temperature is going to be. So a change in latitude is going to change-- explain a change in the average January temperature of US cities. So what I would say here now is latitude is your explanatory variable, and temperature is going to be your response variable. So that is the tutorial about variables. Thanks for watching.
Source: AMERICAN JOURNAL OF CARDIOLOGY
Hi. This tutorial is all about confounding variables. So let's start with a definition of what a confounding variable is. So a confounding variable is a variable that is related to both group membership and the response variable of interest in a research study. Confounding variables have a significant effect on the response variable, which confounds the explanatory variable.
So let's take a look at an example to see if we can make some sense of that definition. So a study that appeared in the American Journal of Cardiology from 2003 found that heart attack survivors who owned a dog generally had better heart function post-heart attack than those who did not have a dog. So the question here is, does a dog help a heart heal faster?
So really, we're trying to see if having a dog causes a healthy heart. So we're trying to figure out, does that relationship make sense? And I don't know. I would probably say that just the fact of owning a dog probably is not going to strengthen your heart. So it would seem that there is a confounding variable at work here.
So dog owners-- so what the article ended up saying is that dog owners tend to get more exercise than non-owners, and exercise helps strengthen hearts. So what's really happening here is instead of making this relationship dog causes a healthy heart, really there's a third variable at work here. And this is the confounding variable is that exercise is actually what causes a healthy heart.
So this relationship is probably not a valid one to make, whereas this one is pretty commonly known, that an increase in exercise causes a healthy heart. So this confounding variable of exercise confounds the relationship of the dog with the healthy heart. So exercise confounds this variable having a dog.
So the level of exercise is a confounding variable that prevents us from concluding that dog ownership improves heart function in heart attack survivors. So it's important that when you are making conclusions of a research study, you want to make sure that there's not really a confounding variable at work if you're trying to show a cause and effect relationship. So that's all for confounding variables. Thanks for watching.