In Psych 144, I always emphasize the importance of distinguishing phenomena (things that we observe) from theories (explanations for those phenomena). For example, there are sex differences in scores on the math portion of the SAT. No question about it. But there are many different ways to explain this fact that might involve biased tests, a biased society, genetics, and so on.
But this blog has nothing to do with sex differences. What it does have to do with, though, is the fact that no one has commented on any of my blogs. That's the phenomenon of interest for today. And in an exercise straight out of Psych 144, I'm going to generate as many theories to explain this phenomenon as I can. I'll worry about which ones are correct later.
1. Students never look at blog.
2. Students do not know how to post comments.
3. Blogs are too boring to comment on.
4. Blogs are incomprehensible.
5. Students are too dense to think up comments.
6. Settings do not allow comments (because blogger is too dense).
7. Students are afraid to interact with instructor.
8. Students are afraid of appearing to be apple polishers.
9. Students find topics uninteresting.
Other suggestions?
Thursday, February 21, 2008
Friday, February 8, 2008
In Praise of Statistical Thinking
In an essay in the Chronicle of Higher education, Eric G. Wilson (an English professor at Wake Forest) writes "In Praise of Melancholy."
Melancholy, by the way, is just a fancy word for depression. The word itself derives from the ancient Greek meaning "black bile," because it was once thought that depression was caused by an excess of black bile in the body.
Anyway, the author of the essay mentions that recent polls show that 85% of Americans rate themselves as happy. He also claims that this is a disturbing trend because melancholy has been the inspiration for much of the world's great art, music, and literature. Yet we are content to "annihilate" it through positive psychology, psychotherapy, and the use of antidepressants like Prozac.
Although there are a lot of interesting things about this argument, there is a fundamental problem that prevents it from ever getting off the ground ... and it's something that I emphasize over and over in Psych 144. In the abstract it's this: evaluating a claim about a statistical relationship requires a comparison of one variable across levels of the other. In the concrete: the author claims that people are happier than they used to be, but he only presents data on how happy they are now. Who's to say they weren't just as happy in the "olden days?" If they were, then the whole idea that we are in the process of "annihilating" melancholy falls flat.
I'm not an expert on happiness (or what psychologists are more likely to refer to as "subjective well being"), but I'm pretty sure there are data out there that show whether or not there has been a change in the happiness levels of Americans across time. I'm also pretty sure that those data show that there really hasn't been much change over time--certainly not changes on the order of what the Wilson suggests in his essay.
But this is a blog, not a journal article. So you'll have to look up those data yourself.
Melancholy, by the way, is just a fancy word for depression. The word itself derives from the ancient Greek meaning "black bile," because it was once thought that depression was caused by an excess of black bile in the body.
Anyway, the author of the essay mentions that recent polls show that 85% of Americans rate themselves as happy. He also claims that this is a disturbing trend because melancholy has been the inspiration for much of the world's great art, music, and literature. Yet we are content to "annihilate" it through positive psychology, psychotherapy, and the use of antidepressants like Prozac.
Although there are a lot of interesting things about this argument, there is a fundamental problem that prevents it from ever getting off the ground ... and it's something that I emphasize over and over in Psych 144. In the abstract it's this: evaluating a claim about a statistical relationship requires a comparison of one variable across levels of the other. In the concrete: the author claims that people are happier than they used to be, but he only presents data on how happy they are now. Who's to say they weren't just as happy in the "olden days?" If they were, then the whole idea that we are in the process of "annihilating" melancholy falls flat.
I'm not an expert on happiness (or what psychologists are more likely to refer to as "subjective well being"), but I'm pretty sure there are data out there that show whether or not there has been a change in the happiness levels of Americans across time. I'm also pretty sure that those data show that there really hasn't been much change over time--certainly not changes on the order of what the Wilson suggests in his essay.
But this is a blog, not a journal article. So you'll have to look up those data yourself.
Friday, February 1, 2008
The Meaning of Imply
While lecturing the other day on correlational research, I had a thought about the famous dictum, "Correlation does not imply causation." No, I'm not going to try to debunk it. But I am going to suggest that it be reworded because the word imply is ambiguous.
One meaning of imply is something like guarantee. This is the intended meaning in "Correlation does not imply causation." The simple fact that two variables are statistically related does not guarantee that the underlying relationship between them is causal. It could be causal, but this is not a sure thing.
A second meaning of imply--which is probably more common in everyday conversation--is something like hint at. So if my wife casually notes how full the trash can in the garage is getting, some might say that she is implying that I carry it out to the dumpster. (Note that I wouldn't say this ... but some might.) So "Correlation does not imply causation" could be interpreted to mean that the fact that two variables are statistically related does not even hint at there being a causal connection between them.
But this isn't right because it does hint at it. For example, if caffeine consumption is correlated with later miscarriages, then this suggests that researchers ought to take a look at caffeine consumption as a possible cause of miscarriages. It doesn't guarantee that caffeine consumption is a cause ... but it could be.
So let me propose a rewording of the old dictum: "Correlation does not guarantee causation."
One meaning of imply is something like guarantee. This is the intended meaning in "Correlation does not imply causation." The simple fact that two variables are statistically related does not guarantee that the underlying relationship between them is causal. It could be causal, but this is not a sure thing.
A second meaning of imply--which is probably more common in everyday conversation--is something like hint at. So if my wife casually notes how full the trash can in the garage is getting, some might say that she is implying that I carry it out to the dumpster. (Note that I wouldn't say this ... but some might.) So "Correlation does not imply causation" could be interpreted to mean that the fact that two variables are statistically related does not even hint at there being a causal connection between them.
But this isn't right because it does hint at it. For example, if caffeine consumption is correlated with later miscarriages, then this suggests that researchers ought to take a look at caffeine consumption as a possible cause of miscarriages. It doesn't guarantee that caffeine consumption is a cause ... but it could be.
So let me propose a rewording of the old dictum: "Correlation does not guarantee causation."
Sunday, January 27, 2008
You CAN Do Anything with a Bachelors in Psych
"You can't do anything with a bachelors degree in psych."
In addition to its being plain incorrect, this claim bothers me for several reasons.
The first is that, according to Tara Kuther in The Psychology Major's Handbook, the majority of psychology majors do not go to graduate school ... ever. But of course they must be doing something.
The second is that it implies that there is something unique about psych majors in terms of their ability to get jobs with only their bachelors degrees ... as though sociology, philosophy, and chemistry majors are being handed jobs left and right. Yes, the technical knowledge obtained by chemistry majors gives them the upper hand for getting certain kinds of jobs, but the technical knowledge obtained by psychology majors (e.g., behavioral principles, statistical analysis) gives them the upper hand for others.
The third reason this claim bothers me is that it reflects a misunderstanding of the traditional purpose of an American-style college education, which is not to be trained for a particular career. Yes there are some exceptions; nursing and engineering come to mind. But a bachelor's degree in psychology is not supposed to turn you into a psychologist any more than a bachelor's degree in biology is supposed to turn you into a physician. What both of those bachelor's degrees are supposed to do is turn you into an educated human being who knows a little about a wide variety of fields, a little more about one specific field, and has lots of important general skills like the ability to write and speak clearly, analyze complex problems, motivate yourself, and work effectively with others.
The fact is that a bachelor's degree in psychology can be one step on the way to doing just about anything--including becoming a psychologist, a doctor, a lawyer, a newspaper reporter, an advertising executive, ....
In addition to its being plain incorrect, this claim bothers me for several reasons.
The first is that, according to Tara Kuther in The Psychology Major's Handbook, the majority of psychology majors do not go to graduate school ... ever. But of course they must be doing something.
The second is that it implies that there is something unique about psych majors in terms of their ability to get jobs with only their bachelors degrees ... as though sociology, philosophy, and chemistry majors are being handed jobs left and right. Yes, the technical knowledge obtained by chemistry majors gives them the upper hand for getting certain kinds of jobs, but the technical knowledge obtained by psychology majors (e.g., behavioral principles, statistical analysis) gives them the upper hand for others.
The third reason this claim bothers me is that it reflects a misunderstanding of the traditional purpose of an American-style college education, which is not to be trained for a particular career. Yes there are some exceptions; nursing and engineering come to mind. But a bachelor's degree in psychology is not supposed to turn you into a psychologist any more than a bachelor's degree in biology is supposed to turn you into a physician. What both of those bachelor's degrees are supposed to do is turn you into an educated human being who knows a little about a wide variety of fields, a little more about one specific field, and has lots of important general skills like the ability to write and speak clearly, analyze complex problems, motivate yourself, and work effectively with others.
The fact is that a bachelor's degree in psychology can be one step on the way to doing just about anything--including becoming a psychologist, a doctor, a lawyer, a newspaper reporter, an advertising executive, ....
Friday, January 25, 2008
Jumping to Causal Conclusions
There's been a rash of studies in the news purporting to show that some medication, food, or behavior (call it X) has an effect on some aspect of health (call it Y). In Psych 42, I've talked recently about new studies on caffeine and miscarriage and anger suppression and mortality (i.e., death). The latest study concerns birth control pills and ovarian cancer.
All of these studies do clearly show an association between X and Y. However, because none of them is an experiment--where the independent variable is manipulated and extraneous variables are controlled--they do not show that X causes Y. And it doesn't matter how many times the journalists or researchers state that they do.
All of these studies do clearly show an association between X and Y. However, because none of them is an experiment--where the independent variable is manipulated and extraneous variables are controlled--they do not show that X causes Y. And it doesn't matter how many times the journalists or researchers state that they do.
In the latest example, the researchers showed that women who took the Pill were less likely than women who did not take the pill to develop ovarian cancer. This could be because they took the Pill ... but it is also possible that women who took the Pill differed from women who did not in some other way. For example, women who took the Pill might also have been women who tended to think about and act on their health and well-being more than women who did not--and this difference might have been responsible for the reduction in ovarian cancer.
Yet the article linked above is titled "Pill Prevents Ovarian Cancer for Decades." The word "prevents" clearly implies causation. But if the risk reduction the researchers found was due to something other than the Pill, then then it is not the Pill that is doing the "preventing" and going on the Pill will not reduce anyone's risk.
There are other issues here--such as the effectiveness of statistically controlling for confounding variables--but I'll cover those in another blog. For now, let me give my standard disclaimer about this sort of thing. I am not saying there is anything wrong with this research. It is interesting and important and tells us that there is an association between X and Y that might be a causal one. But researchers, journalists, and the general public need to be much more cautious about jumping to causal conclusions.
Yet the article linked above is titled "Pill Prevents Ovarian Cancer for Decades." The word "prevents" clearly implies causation. But if the risk reduction the researchers found was due to something other than the Pill, then then it is not the Pill that is doing the "preventing" and going on the Pill will not reduce anyone's risk.
There are other issues here--such as the effectiveness of statistically controlling for confounding variables--but I'll cover those in another blog. For now, let me give my standard disclaimer about this sort of thing. I am not saying there is anything wrong with this research. It is interesting and important and tells us that there is an association between X and Y that might be a causal one. But researchers, journalists, and the general public need to be much more cautious about jumping to causal conclusions.
Monday, January 21, 2008
From "Grand Theft Auto" to Inferential Statistics
A Research Methods student from last semester, Jordan Prendez, sent me the following link: http://www.gamespot.com/news/6147420.html. It is a short article describing a research study showing that players of a more violent video game are more prone to violence themselves than players of a less violent video game.
First, let me point out that the study, as described, is a good example of a two-group randomized experiment with multiple dependent variables. It is an experiment because it has a manipulated independent variable and the researchers appear to have tried to control other variables.
Second--and this was Jordan's point--it is interesting how the gamers who commented on the article attacked the research (presumably without reading the original study). This is reminiscent of a famous study by Lord, Lepper, and Ross (1979), showing that people are good at finding the flaws in a research study ... but only when they already disagree with the study's conclusions.
Statistics students, in particular, should note the argument that appears in a couple of the comments. In essence, "There's no way that a sample of 100 [the number of participants in the study] can represent the whole population of gamers." Intuitively, this seems right because 100 is a very small fraction of the total number of gamers out there. If there were 10 million gamers in the population, for example, this sample would only represent 1/100,000th of that population. But this intuition is wrong. A sample of 100 will generally be much more similar to the population--even a very large population--than most people realize. And this is what the field of inferential statistics is all about.
Thanks, Jordan!
First, let me point out that the study, as described, is a good example of a two-group randomized experiment with multiple dependent variables. It is an experiment because it has a manipulated independent variable and the researchers appear to have tried to control other variables.
Second--and this was Jordan's point--it is interesting how the gamers who commented on the article attacked the research (presumably without reading the original study). This is reminiscent of a famous study by Lord, Lepper, and Ross (1979), showing that people are good at finding the flaws in a research study ... but only when they already disagree with the study's conclusions.
Statistics students, in particular, should note the argument that appears in a couple of the comments. In essence, "There's no way that a sample of 100 [the number of participants in the study] can represent the whole population of gamers." Intuitively, this seems right because 100 is a very small fraction of the total number of gamers out there. If there were 10 million gamers in the population, for example, this sample would only represent 1/100,000th of that population. But this intuition is wrong. A sample of 100 will generally be much more similar to the population--even a very large population--than most people realize. And this is what the field of inferential statistics is all about.
Thanks, Jordan!
Sunday, January 20, 2008
Why Indeed?
I guess this blog started out as a test. I wanted to have students in my Intro to the Psych Major course (Psych 60T) create and maintain blogs for short, informal, reflective writing assignments. So I had to create one myself to see how it worked. Then I had to try posting to it. Then I realized it would be a good forum for communicating with my students, especially about ideas that in class would be considered "digressions." (I'm a big digressor.) I suspect, however, that these ideas are probably more interesting and useful than the "course material" in many cases. I also realized that I could do this without cluttering up the Announcements space on the course Blackboard site ... and I could make these ideas available to students in all of my classes ... and rest of the world too. Welcome to the 21st century, Dr. Price.
Subscribe to:
Posts (Atom)
