Continuing and concluding our discussion of assessment and research practices in personality theory, we now come to the clinical interview. Typically, a clinical interview takes place over an extended period of time and goes into either a surface level, a medium level, or an in-depth survey of the individual who is being interviewed. It's referred to as a clinical interview typically because it takes place in a clinical setting between a therapist and a patient.
But the clinical interview can also be used and is also used for in-depth research between a subject and a researcher. The subject or the interviewee will come into the office and go through an extended period of in-depth interviews. From that, information is gathered and contributed to the research project.
If you're interested in clinical interview technique, there's an incredible book that investigates relationships. It's called Love and Limerence by Deborah Tennov. And Tennov uses the clinical interview technique not as a therapist, but as a researcher.
This is a research psychologist who interviewed individuals regarding their experience with falling in love and being in relationships. And through a systematic series of interviews with many different individuals, hundreds of different individuals, she found insight into what she refers to as love and limerence, or relationships and crushes. So this is a fine example of an entire book that's based on clinical interviews.
Behavioral assessment is another technique that's used, and this is when an operational definition is set. In other words, a trained professional observes someone's behavior and makes note of describing the individual's behavior. And that is called behavioral assessment.
Now, thought and experience samples. This is a very common method used in personality psychology for research, and it's really a self-assessment. At different moments, the individual record what they're thinking, what they're feeling, and keep a log of these thoughts, feelings, and behaviors, and actions. And this is then kept as data, research data. This is referred to as thought and experience samples. So it's really an introspective phenomenological approach to research. These are all valid forms of doing research into personality.
But let us turn now to our two final quantitative methods of research, and that's experimentation and correlation methods. Experimentation has been probably the predominant and preferable choice of most scientific and quantitative psychologists. And this is because it offers the cause-and-effect evidence for a cause-and-effect relationship. Other forms of research don't provide this.
There are limitations to this benefit of establishing a cause-and-effect relationship, and that is that not all circumstances in the human condition are observable and measurable under experimental conditions. Not all qualities of the human condition are observable and measurable, and especially, it seems, things such as personality. How we observe and measure personality is still quite a controversial mystery.
Even the concept of cause and effect necessitates the idea of determinism, which we discussed earlier in this lecture. And we'll find, especially in theories of personality, that cause-and-effect thinking is not very popular amongst most theorists in psychology.
By contrast to cause and effect, there is the concept of overdetermination. And that's the idea that multiple factors influence an individual phenomenon or an individual cause. So rather than having cause and effect, we have causes and an effect.
So keeping that in mind, the experimental method, as a review from what you have studied in an introduction to psychology course at least, is the controlling of all variables except one variable in a situation. And in so doing, we determine whether that one singular variable, when either present or absent, changes the results of the experiment.
So we have what's typically called a control group and an experimental group. And each of those groups are identical in all ways but one variable. And if there's a difference in the results between the two groups, then it must be concluded, through statistics and through logic, that there's a cause-and-effect relationship between that one variable and the change that took place.
So let's take an example. Let's say, for sake of an example, there was a research study that a student once performed in my course in social psychology, and they were interested in how a red cup might affect users of a social media website to say yes to their friend request.
So they made two identical profiles. Everything in the two profiles were completely identical, with the exception that in one profile the profile picture showed an individual, a male, holding a red cup, a party cup, a Solo cup. And in the other, it was just the individual, same individual, standing there. Everything else in the photo was identical, the absence of the cup.
And as it turned out, after sending out numerous random friend requests to individuals, there was a difference of 80% of the individuals who were friend requested with the red Solo cup profile said yes to the request as opposed to only 20% of the group without the red cup. So we might be able to conclude from this that there's a cause-and-effect relationship, that having a red cup in a profile picture on a social media network causes more individuals to say yes to a friend request.
Now, unlike classical Newtonian mechanics of billiard ball interactions and laws of physics, we can't conclude that 100% of the time the red cup causes this fixed action behavior of responding yes to a friend request. That's obviously a silly conclusion.
So when we are talking about these cause-and-effect relationships in psychology, we are always talking relative to the individual. 80% of individuals is certainly not 100% 100% of the time. So I caution students to always take cause-and-effect research for what it is concerning cause-and-effect relationships or any sort of lawful human behavior while psychology is still looking for its first law of human behavior.
So regarding cause-and-effect relationships and experimentation, we typically have a control group and an experimental group. An experimental group-- there can be more than one experimental group. You could have a control group and simultaneously run as many experimental groups against that control group as one wishes.
But in the simplest form, there's one experimental group, one control group. Each group has at least 40 participants to make it statistically valid. That's a lowercase n of 40 in each group. And in total, you'd have 80 subjects. Large uppercase N is the subjects of the entire study, about 80 to make it valid statistically.
Random sampling is used to select the participants for each group randomly. That means anyone in a given population has equal chance of being assigned to either group or even being assigned to the study.
Correlational research does not show cause and effect. When we look at correlational research, we're looking at relationships between two or more variables. Now, if there is a strong relationship between two variables, it doesn't necessarily exclude cause and effect. But using correlational methods, we cannot claim cause and effect as we might be able to in experimental research.
So in a correlational research, you simply have two separate variables, two things, and the occurrence of those two variables together is measured. So for example, we can have a linear correlation, such as the correlation or the relationship between SAT scores and college freshman GPA.
It turns out that there is a very positive, highly positive, correlation between what one scores on their SAT while in their senior year of high school and what their GPA will be in their freshman year of college. So there's a positive correlation, meaning as one goes up, the other goes up. This doesn't mean that taking the SAT causes the high GPA a year later. What it means is that it has predictive value.
Now, again, a strong correlation, a strong positive correlation, positive doesn't mean good or bad. Positive certainly means that as one variable increases, so does the other. So in a positive correlation, we can have a very strong correlation. That means it's very strongly correlated that as one goes up, the other goes up.
And we could also have weak correlations. And that means as one goes up, the other might go up slightly or a little bit, but not dramatically. So we have different levels of strength in our correlations.
We also have negative correlations. And that means as one goes up, the other goes down. So in a negative correlation, as, say, time partying goes up, GPA goes down, that's an inverse or a negative correlation. Negative does not necessarily mean bad. As one exercises more, their health issues decrease. Now there is a negative correlation. So when we talk about negative and positive, we're simply describing the relationship between two variables.
So most of psychology research is actually correlative research. And certainly when we're looking at personality theory, a lot of the things that we discuss are correlational. As a matter of fact, the majority, as I said, of psychological research does turn out to be correlational.
And although it's not praised as much as experimental design, and there are philosophical and political reasons within the science of psychology for why that exists, which you'll study that more in a history of psychology or a philosophy of psychology and science class. But for right now, we should keep in mind that correlation is a very useful predictive tool.
Now, in correlations we can have what are known as linear correlations. And that is there's a direct relationship, either positive or negative. As one goes up, the other goes up, or as one goes up, the other goes down, the variables. That's a linear correlation.
And we also have non-linear correlations. And non-linear correlations are things including a U curve or an inverted U. So in a U curve, that means that variables will go down to a certain point, and then they'll begin to rise again. So in other words, there's a certain point at which the relationship changes.
And a common example of this would be a U curve is typical in talking about years of marriage and report of enjoyment and satisfaction in those marriages. And it turns out about 20 years into the marriage, the marriage begins at honeymoon as very enjoyable, and then after a few years enjoyment level is reported to drop, till about the 20th year when it starts to rise again.
And so that would be a U curve, and this is non-linear correlation. And some of the explanations for this could be the child leaving home and the couple back together just the two of them again after the children leave the house.
An inverted U is a commonly described non-linear correlation between income and enjoyment of life or satisfaction with life. It turns out that to a certain degree, income will increase with the pleasure of life. And at a certain level, income and pleasure of life reverts back down, and it's typically around $70,000 a year. That income in excess of $70,000 a year, individuals start reporting a decrease in the enjoyment of life. And that, if you could imagine, looks like an upside down U on a correlational graph.
So we have non-linear correlations. We have linear correlations. And they're simply showing relationships to one another rather than necessarily a cause-and-effect relationship.
Finally, I'd like to make a few remarks regarding the scientific method. The scientific method is unique in the way that it's employed in psychology. The scientific method itself is something that changes from one science to another. How science is done is dramatically different in physics as it is in biology, chemistry, to social sciences. So we should keep in mind that this idea of a scientific method is specific to psychology.
But the thing that is common within scientific method is that it really comes in three stages. We can understand this in three stages. And the three stages are not unique to science. They are identical, whether we're doing philosophy, or whether we are doing theology, or any type of endeavour where we ask a question, we come up with an answer, and we find evidence for that answer to support that-- to support that argument. So we're really talking about like a court of law of ideas.
So as the scientific method in psychology goes, we state an idea, a solution, some sort of idea of why things occur as they do or a solution to a problem. We call this a hypothesis, a hypothesis. So the hypothesis, the hypothesis, is something that is hypo. It's below thesis, the theory. It's not yet a theory. It's an idea.
We then look for evidence to support that hypothesis. And that evidence in psychology is found through the methods that we've just explored in the past lecture. That could be the clinical study method, the correlational method, the projective test method, the experimental design method, and the behavioral observation method. So we look for evidence to support or disapprove our hypothesis.
And then if we have successfully either supported it or not disproved it, it then becomes a theory. So the idea is that there's hypothesis, hypothesis testing through research, or finding evidence, and then presenting the theory.
The next step is when a theory is so regular that it never changes and one can always count on it. That's when it becomes a law. And so far in psychology, we have no laws. There are laws in psychophysics, such as laws of how sensory organs function. But that's physiology. That's not psychology. That's a branch of physiology.
When we talk about psychology and certainly personality theories, we have no laws and very, very few strong cause-and-effect relationships between anything. Most things are correlational.
So those are some ideas to keep in mind about research methods, about the scientific method as it's used in psychology. And I think now we are ready to approach our first theory of personality, which we'll begin Lecture 2, Week 2, which is the foundational psychoanalytic theory of Sigmund Freud.