Latest / Professor Peterson’s Psychology Classroom / Inside Psychology Research: From Clinic to Lab

Transcript

Inside Psychology Research: From Clinic to Lab · Professor Peterson’s Psychology Classroom

0:05
Welcome to Professor Peterson's Psychology Classroom,
0:09
connecting the science of human behavior to everyday
0:12
life. Here's Dr. Curtis Peterson.
0:17
All right, so this section is on research methods, and,
0:21
uh, the point of this section is just to provide an understanding
0:24
how psychologists go about
0:29
investigating behavior and mental processes.
0:32
And as I begin this talk, I, I want to make it clear that all
0:36
psychologists use this method. Okay, sure,
0:40
experimental people test, test, test,
0:43
but even in applied psychology, we use this methodology
0:48
to try and develop an idea of a
0:53
situation. And I'll give you an example. These— let
0:57
me back up. These are our basic objectives of today,
1:01
is that we're going to review the basic research methods, different research
1:05
methods used in psychology, and start to develop some research proposals,
1:10
maybe some ideas of what they'll be.
1:14
So when we talk about investigation,
1:18
and I'm going to go back to an idea here.
1:22
Is that when,
1:30
when we look at behavior,
1:34
we start with a research question. And as I said,
1:38
applied and research psychologists do this.
1:42
So in the applied area, I'll look at clinical psychologists.
1:46
That's people who, who treat, assess and treat mental
1:51
disorders. And in the experimental psychology
1:54
area, I'll use a social psychological investigation.
1:59
So for example, a research question when you're a clinician is,
2:04
does this person have depression? Okay, and we'll go through
2:08
the clinical one. So a person comes in,
2:11
they say they're feeling depressed, they feel they're doing
2:16
things. So what do I do? So the research question,
2:20
is this person depressed? Well, what I need to do
2:24
is investigate. So in the
2:28
clinical setting, the, the, the client becomes kind of what
2:32
we're calling this literature review, where we give them
2:36
some tests, we give them some interviews,
2:42
uh, we, we do assessments of all, all kinds depending
2:46
on what it is, and we form a hypothesis.
2:51
So let's say after doing the review of
2:54
the client, I do hypothesize that I—
2:57
this client has depression. Okay,
3:01
what I then do is I implement a method
3:06
to alleviate that depression. This method
3:10
usually is called the treatment, the treatment that we're going
3:13
to have the client do. So we're going to put the client through
3:17
some treatment modality, but during that treatment modality,
3:21
we're going to measure their qualitative experience.
3:25
Okay, do they see their depression going down? Do they,
3:29
do they start to feel the symptoms be alleviated?
3:32
But we're also hopefully going to measure it as well. We're going
3:36
to give them— every time they come to our office, we'll give them a
3:40
quick, short Beck Depression Inventory, which is a very short
3:44
8-question, uh, down-and-dirty assessment so
3:48
that I know both objectively and subjectively
3:52
whether or not my client is getting well.
3:56
This is using a qualitative and quantitative method This
4:00
is called a mixed method.
4:04
So, and then what I do is at the end,
4:08
I reassess. All right, I go back to my
4:12
research question. This client had depression, and then I'm going to
4:15
interview the client again. I'm going to do more assessments to see if
4:20
my method, my treatment worked. Okay,
4:24
so this is a clinical process. What about the research
4:28
process. All right.
4:35
So the research process, we'll use
4:39
a black pen. We start with a research
4:43
question. Okay.
4:46
So let's see,
4:49
what is a common research question that I see from the
4:54
research we read. Um,
4:59
the, the one that is in the, uh, the, the classroom about
5:03
moral identity, okay? Um,
5:07
and, and does it influence a person's
5:11
cheating behavior? So in, in that one, in the— and you'll
5:15
find it in the, um, the, the reading section
5:19
of the class which is up next,
5:22
is does activating
5:27
one's identity of being
5:30
moral influence
5:39
whether or not the individual will cheat or not? Okay,
5:43
and that's the basis of this, and I encourage you to go through
5:47
that article that is posted.
5:50
Identity is what my research question is, and I want to know how
5:54
identity influences behavior. In this case, we're going to use
5:58
our— an individual's moral identity. So the investigation—
6:03
so I have 2 concepts that I have to investigate.
6:06
I have to investigate whether or not identity
6:10
has been explored in this way. What is identity?
6:14
What are the aspects of it? What is the theory behind it?
6:18
And then I have to explore the psychology of morality.
6:23
And then I need to make a connection logically
6:28
between identity and morality, because that's
6:32
what's going to run my hypothesis. My hypothesis
6:37
being that if one's moral identity is activated,
6:41
then that will influence an individual's behavior.
6:45
Now, what kind of behaviors do we believe is moral?
6:49
Well, not cheating, right? So I'm going to put C for cheating.
6:53
So what we have is we have an activation of
6:56
identity which
7:01
should produce a moral behavior,
7:06
which then should reduce unmoral behavior such
7:09
as cheating. Okay. So I'm going
7:13
to develop a research plan. In this one,
7:18
what I need to do is I need to compare at least
7:21
2 groups. Now I'm simplifying it here.
7:25
When you read it in the module classroom,
7:29
there's a lot more detail. He actually has 6 hypotheses,
7:32
but I'm doing simple example right here.
7:37
What I need at least for this research is 2 groups.
7:41
One where I activate one's moral identity and
7:47
one where I don't activate their identity.
7:51
So I'm going to put a 0 representing this.
7:54
This is going to become the experimental group. This is going to become
7:58
what's called the control group. Okay.
8:02
Then what I need to do is I
8:06
need to put a
8:13
research— I need to come up with a situation where
8:18
a person has the opportunity to cheat.
8:22
So I'm going to activate a moral identity. In this case,
8:26
I believe he had someone, the participants
8:29
read about someone acting moral and virtuous.
8:33
And then in his control group, the researcher
8:39
had them read something completely unrelated to identity.
8:42
I think it had something to do with the sea, ocean, and ships
8:47
and something like that. But then I need to put both groups
8:51
in a condition where they have the opportunity
8:55
to cheat. Okay, so this is going to be our C
9:00
condition, cheating condition, right? This is
9:05
our independent variable, meaning this is what we're going to manipulate,
9:10
and we can represent that with an I. I know this might get confusing because
9:13
I use that as identity. This is
9:17
our dependent variable because we believe
9:21
the behaviors that are committed here are dependent upon
9:26
our independent variable. Additionally,
9:29
our dependent variable is what we're going to measure.
9:33
Okay, and then we're going to then look at the
9:38
results. All right,
9:43
and again, what, what, what he found is, is that
9:46
when you activate someone's moral identity, they are less
9:50
likely to cheat. Hence, identity,
9:54
when it's activated, will produce the behaviors
9:58
that that identity expects out of the person.
10:01
But you'll get more in depth when you read this in
10:05
the classroom. So, but this is the basic process.
10:10
Now what I want to talk about quickly, even,
10:13
and it's expressed in the classroom
10:18
best, so I'm just going to summarize it. This is what
10:22
is the difference between qualitative, quantitative, and mixed models.
10:26
Simply put, mixed models— and I'm going to simplify this right now because
10:30
there's no other reason to, to do it— is mixed models means you do both
10:34
a qualitative method and a quantitative method. And I'll explain in some conditions
10:39
why you would do that. But let's go on. So this
10:44
interesting graph, it breaks down the difference in quantitative
10:48
and qualitative situations.
10:52
When I'm doing a quantitative study,
10:56
what I am wanting is the
11:01
objective reality
11:05
of whatever it is I'm studying. Okay,
11:08
so I'm looking for, in this case,
11:12
numbers,
11:13
objective measurement. Of some
11:17
type. When I'm doing a qualitative
11:21
method, on the other hand, what I want to know is
11:24
the subjective reality of an individual.
11:28
I want to know your reality versus the reality of
11:33
your larger group in which I'm trying to explore.
11:38
So I'll give you an example of a mixed method before
11:42
we go on. So when I used to do organization
11:45
consulting and an organization that was in crisis would call me
11:50
in and they'd ask, hey, we're not sure what to do,
11:53
can you help us out? Usually what I start with
11:56
is a quantitative method. So usually it's in the form
12:01
of a survey, many of the times surveying job satisfaction,
12:05
job burnout, job commitment.
12:09
Satisfaction with a whole bunch of things to try and get an understanding
12:14
of that organization culture and the people that work within that
12:18
organization. But that only gives me this broad picture,
12:21
right? This, this kind of photo shot of what's
12:25
going on. What I don't know is— so
12:29
I have these objective results that maybe this organization
12:33
has a high rate of burnout. What I don't know is
12:37
what is the subjective reality of those
12:41
people working in an environment that
12:44
is— that has a high burnout rate. So at that
12:49
point, I switched to a qualitative method where I'm going to go interview
12:52
employees. I'm going to say, you know, based on my results,
12:56
this is what I found. Can you tell me what your experience is?
13:00
And then after that, again, going back to this,
13:06
I'm going to implement a treatment, right?
13:09
I'm going to provide the organization with some training,
13:13
something, and then I'm going to go back and remeasure to see
13:18
if there were any changes in the organization's behavior.
13:22
With that being said, we've explained what a qualitative
13:26
method is— subjective reality. Quantitative method is objective.
13:30
Mixed methods is when you really need to have a strong understanding
13:35
of both the objective and subjective experiences of people within
13:39
a given group or within a given society.
13:44
When we break down quantitative and qualitative methods,
13:48
quantitative methods is broken down into
13:52
2 basic categories. One is
13:55
descriptive. The second is experimental.
14:01
When we talk about descriptive methods, what we're talking about
14:04
is just describing a given population,
14:09
a given group, or a given— whoever we are
14:13
researching. Its means,
14:16
meaning averages, and deviation,
14:19
meaning how How closely do people measure with
14:23
each other? It's bar graphs,
14:27
it's pie charts, it's line graphs. It's all of
14:30
those things that give us a photo picture
14:35
of the population that we're looking at or the
14:38
variables that we measured.
14:40
Okay. When I use the word
14:44
variable, what I'm talking about is independent and
14:47
dependent variables. So in this instance, they Our
14:52
subject matters are usually always human.
14:56
Whoops. And our independent variables
15:00
are what we manipulate. So this is a group, individual activation,
15:04
non-group activation. And this is our measured variable,
15:07
measure of cheating behavior. So when we talk about variables,
15:11
that's what is meant. The other
15:15
things that descriptive methods try to get at
15:19
is whether or not variables are associated
15:22
with each other. And there's 3 kind of main methods
15:26
to do that. There are more kind of out there, but the most
15:30
common ones are correlation, regression, and factorial analysis.
15:35
Correlation is when I just want to know if 2
15:40
things are associated with each other. So as one
15:45
variable goes up, does the other variable go up? And vice versa,
15:48
when one variable goes up, does the other variable go down?
15:53
So, um, an example of,
15:56
uh, of correlation in, in, um,
16:01
uh, in, in psychology is that there is a
16:05
correlation between altitude and
16:08
suicide rate in that The
16:12
higher the altitude, the higher the suicide
16:16
rate. Okay.
16:21
I bring this example up because
16:25
one, we're talking about descriptive methods.
16:29
And this is very important. Descriptive methods cannot
16:34
determine causation. Okay. Because we have
16:38
this simple association. As a person goes up in altitude,
16:42
the risk of suicide increases, right? But is
16:47
being on top of a mountain causing suicide
16:51
or someone to be suicidal? And I
16:55
will tell you, there's many other maybe causal variables.
16:58
There's been the hypothesis that it is the
17:03
thinning of the air. Um, the most
17:06
viable hypothesis, as is if you increase in altitude,
17:10
you become more and more disconnected with other human
17:13
beings. And we know from further research
17:18
in suicide, the most common variable in
17:22
suicide, the most common variable is either
17:27
human rejection, loss of a relationship,
17:31
Loss of a job, which includes people,
17:36
and loneliness and isolation. Okay, that those—
17:41
if you think about those, all of those variables are about being socially
17:45
disconnected from other people, and that is one of the,
17:49
the most common, uh,
17:53
reasons people commit suicide. So So
17:57
what I'm saying is, is from correlation, we cannot determine
18:01
causation. All we can say is that these 2 things are associated
18:05
with each other. When we get into regression,
18:09
what we want to know is, does one variable predict
18:14
another variable? So in this example,
18:18
we look at, does— the
18:21
most common that we use in introduction to psychology classes
18:25
is does study time influence scores
18:30
in exams? And yes, we do find
18:34
that there's a correlation between the number of study hours and the score on exam,
18:38
and that's an important association to keep in mind, please.
18:42
And we also know that study time predicts
18:47
test scores using regression analysis.
18:51
The nice thing about regression analysis is that I can actually take it
18:55
a step forward. Yes, I can say one variable predicts the other,
18:59
But I can also determine the— what's
19:03
called the effect size. For regression,
19:07
the effect size is the percentage of one
19:11
variable, in this case exam scores,
19:16
are explained by another variable,
19:19
in this case study time. And in this
19:24
instance, we know that test scores— 30%
19:30
of test scores can be explained by the number of hours
19:33
that a student studies for that exam, which is significant,
19:37
right? Meaning that 30% of what
19:41
your test score is going to be is going to be totally based on sitting
19:45
down with your information and
19:49
reading through it and studying it.
19:53
However, the problem is, is yes, so we can time it,
19:57
but what about that other 70%?
20:01
What else is involved in study
20:05
time and anything else that would
20:10
explain the other 70% of test scores? That's regression
20:14
analysis. I'm going to use that— I'm going to leave factorial
20:18
analysis out for now. I will explain it when we get
20:22
to biological psychology and social psychology
20:26
because it's a little bit more complicated than this introductory example.
20:34
The experimental design is when we are looking
20:37
for cause. And effect.
20:42
Okay, and this is where
20:46
we just explained this experimental design in the identity
20:50
versus no identity. So we manipulated
20:54
an independent variable, we had 2 groups,
20:59
the, the non-identity group was more likely to steal
21:03
than the identity group. So we know that a moral
21:07
identity causes a change in
21:10
behavior. So,
21:14
and that change in behavior is the reduction
21:18
in cheating behavior based compared to
21:21
a control group. Okay, that's the experimental
21:25
design. When we get to qualitative,
21:28
meaning again subjective reality, we do There's
21:33
3 basic types of analysis. We do one on the individual level,
21:37
the group level, and a cultural level.
21:40
Okay? So this may be a racial group, it may be
21:44
a very specific group. Individual, of course,
21:48
the best thing of this would be like a case study
21:52
where a case study is an intense investigation
21:58
into a single person. So someone's biography, would be a good
22:02
example of a case study. Narrative analysis
22:06
is when we, when we want to do a case study on someone who has
22:10
passed away. So we go back in newspaper archives, journals,
22:14
what other family members have written about that person, and develop a notion
22:19
of who that individual is. And then we have
22:23
interview and observations. This is both for the group level
22:26
and the individual level where And these are really
22:29
more techniques than they are—
22:34
more like up here where these were tools. These are kind of tools that we
22:38
use. So intense interviews, observation, that's common with children.
22:43
And then when we get to the group and cultural level, we add
22:46
what's called grounded theory. Grounded theory approaches when I take
22:50
a theory, let's say a
22:59
theory of racial bias of some type.
23:03
And so I take that theory and I go in a community
23:07
of some— that there tends to be a lot of
23:10
racial discontent, and I test
23:14
that model through using interviews and observation and
23:18
information that I collect from that community.
23:22
On the cultural level, psychology is getting into this
23:26
more and more with the study of cultural psychology.
23:29
But this is really more of a sociology
23:33
and anthropological approach, so I'm not going to get into it
23:37
much, even though I do want you just to understand that
23:41
psychology does do cultural analysis— analysis,
23:45
I apologize— but it's usually what's called
23:49
comparative analysis, and that gets into more of these quantitative
23:53
methods. So these are the different methods of analysis.
23:58
And then when we get into, you know, we want to make sure whatever we
24:01
do in the lab also maybe has a field representation,
24:06
and we want to make sure it's congruent with what we observe in reality.
24:10
And this, this is what, what creates our situation.
24:14
So Much more in-depth information
24:18
in the classroom.
24:22
And, and like I said, I'd like this section to go really quickly.
24:25
So we'll only spend about a week on it. Make sure you just have the
24:30
understanding of the different methods that scientists use.
24:34
And also remember the limitations, because again, the majority of
24:38
studies in the field of psychology are correlational and regressional. And we should
24:43
never take that as cause and effect. We should only take that as there's
24:47
relationships between variables. One of the common mistakes I
24:51
see in people's interpretations of psychological data is
24:56
they take a correlation and they assume that there's
25:00
some kind of cause and effect between those 2 variables.
25:03
And most of the time, we're wrong.