Comprehensive Guide to Research Approaches in Psychology

the topic of this lecture will be approaches to research so let's start by talking a bit about

the scientific approach so what we're assuming our basic assumption that we're making in the area of psychology is and

in all areas of science is that events are governed by some sort of lawful order right there's some underlying

structure or laws that we can potentially understand with enough research and observation I

So within psychology our goals are going to be first of all to describe what's going on right to describe the behavior

the mental processes that we're actually observing once we've have a decent description of

what's going on then our hope is to be able to explain and predict those sorts of behaviors and mental processes

as we talked about previously we use things like hypotheses and variables a hypothesis as we've already talked about

is going to be a tentative statement about the relationship between two or more variables

now variables or which we've also mentioned in the past but I want to make sure that we're all on the same page

when we're talking about a variable a variable is going to be anything that's measurable it can be a condition it can

be an event it could be a characteristic or a behavior that are either controlled or observed in a study

the important thing is that variables are free to vary they're free to change and differ either between individuals or

between situations or within the same individual or situation over time so for example we could have things like height

which is a variable that could change that could differ between individuals right so for example I am six feet tall

right other people may be taller or shorter than I am right so height would be a variable

also height can change over the course of the lifespan right so for example if you're uh if you're measuring uh

children during uh late childhood and Adolescence right so for example you might start measuring children when

they're 10 years of old 10 10 years of age see how tall they are at that age then measure them again one year later

at age 11 then one year later at age 12 right in those cases height again is a variable now it's able to change within

the same individual as well as changing between individuals like differing between individuals as well

if we uh describe something adequately if we explain and we're able to predict it we can also then potentially engage

and control sorts of processes right so for example if we understood if we describe the process of depression right

what is depression if we were able to explain it if we're able to predict it in some way then we would have some

better chance of exercising some level of control for example developing uh treatments for depression things like

medications or cognitive behavior cognitive behavioral therapeutic techniques that might help alleviate

depressive symptoms in individuals who are experiencing depression we've talked a little bit in a previous

lecture about the role of theory construction but I just wanted to return to this because it's such an important

issue so first of all if we kind of jump into this process here with the theory right

so again a theory is going to be a coherent network of explanatory ideas basically it's going to be our way of

understanding the way the world operates at least some part of the world once we have a theory then we may develop

hypotheses right these specific predictions that are derived from our Theory right so we develop a theory then

the hypotheses are typically going to follow from the theory once we have our hypotheses then we can

actually develop Empirical research we can actually go out and conduct a study right that's intended to test right the

hypotheses that we've developed now once we've conducted our study right if our findings support our hypotheses

right then confidence in our theory is going to increase right one of the examples we used previously when

introducing the idea of theory and hypotheses is we talked about the idea that people may develop a theory that

depression comes from biochemical imbalances right so for example if that's our theory about where depression

comes from and why it exists we could then develop specific hypotheses so if we think depression is due to

biochemical imbalances we could then Target and develop hypotheses about specific neurotransmitters for example

like serotonin we could develop a medication that may impact the level of serotonin in the brain and see if when

we impact that level of Serotonin by making it go up in some individuals does that alleviate their depressive symptoms

so we could then conduct research to test that if we find support for our Theory right

if our if our research findings support our hypothesis right then that may actually increase our confidence and our

Theory right so if we have an experimental group who gets a drug that increases the level of serotonin in

their brains if we have a control group that we're comparing them against it doesn't get the same drug right rather

they get a placebo a sugar pill right that doesn't have any actual active ingredients right if the people getting

the real dose of the drug are now reporting lower levels of depression right if they're feeling better right

than the control group this is going to increase our confidence right in our theory that there's some sort of

biochemical basis of depression now in contrast if we conduct our study right and the findings do not support

our hypotheses right so let's say we go out we execute our study we've now tested to see whether or not increasing

the level of serotonin has an impact on depression and we find that actually the people taking the the drug to increase

their levels of Serotonin don't look very much different at all right from the people in the control group who got

the sugar pill right this may actually decrease right a confidence in our Theory right because this isn't

consistent with what we expected now if this is the case we now have a couple of options one option is we could

completely discard the theory right we can decide that up maybe there isn't a biochemical imbalance at the at the

heart of depression after all or the other possibilities we may want to go back and revise and refine right our

Theory right so maybe they're maybe there's still a biochemical imbalance but maybe something just kind of went

awry in the research process there'll be a couple of things we could consider one is maybe we targeted the

wrong neurotransmitter right so maybe we're right there's a biochemical imbalance but maybe it's not about

serotonin or maybe the medication that we were using to Target serotonin either wasn't doing that very effectively it

wasn't doing it in the right areas of the brain or maybe it was having some other unintended side effects right so

there are a number of things that we may want to go back and kind of rethink our theory if our results are not consistent

with what we were expecting right and again one of the important things to consider is that this is an

iterative process meaning that scientists go through the cycle again and again and again hopefully building

on and improving what we understand about the phenomena that's being the phenomenon that's being studied

so some basic terminology that you'll need to be familiar with over the course of the semester uh first we'll talk

about things like operational definitions when we're talking about an operational definition what we're

talking about here is going to be exactly what it is that we we mean by each variable so for example uh going

back to our previous example about depression if what we're going to be interested in is seeing how depressed

someone is we have to operationally Define exactly what do we mean by depression

so for example one approach not certainly not the only approach but one approach we could take is we could give

our our participants in our study we could give them a self-report measure of depressive symptoms there are a number

of them that have been developed over the decades but one of the one of the most common ones is something called the

back depression inventory it's a self-report measure that asks people about various symptoms of depression and

asks them to to choose responses indicating to what extent they've experienced each of these different

types of symptoms and so that could be our way of operationally defining depression for our study right now other

people might come along and conduct another study about depression and they may use a slightly different different

operational definition right and that's okay but what's important is that they have to clearly Define right how it is

that they're measuring depression right in their particular study and that's what we mean by an operational

definition the next term that we need to be familiar with is the idea of

participants or subjects right and these are going to be the organisms whose Behavior were systematically observing

in a study um one of the one one of the basic conventions in Psychology is we almost

always refer to uh humans as being participants in studies and when we're studying non-human species like dogs or

cats or gorillas we typically refer to them as being subjects right and that's simply a convention in the field

there are a number of data collection techniques that we'll be talking about as the semester unfolds and these are

going to allow us to engage in empirical observation and measurement we'll talk about some of those techniques over the

over the rest of this lecture and other lectures for chapter two we also are going to be referring to

statistics which are going to be used to analyze data and to decide whether our hypotheses are supported or not right so

it isn't the case that researchers simply decide kind of on a whim whether or not

their hypotheses have been supported rather they're we there's often the case that we use statistics which is going to

be these mathematical operations that we use to decide whether there are differences that are large enough for us

to consider those differences to be real between groups for example so going back to our previous example

about giving people who are depressed a dose of a drug that we think is going to increase their serotonin levels or dose

of a drug that's a placebo that we don't think is going to have any active impact on serotonin we could then use a

statistical analysis to see well after we've given them their drug for let's say six weeks enough time for the drug

to actually be in their system and start having an impact on them we could then measure their self-reported depression

levels to see is there a large enough difference between these two groups for us to say that the difference is real

right that it isn't simply due to chance fluctuations and and the levels of depression between the two groups and

we'll talk more about statistical analyzes as we go through the rest of the semester

what I want us to spend a little bit of time on now is going to be talking about different research methods that are used

in Psychology and so what we're going to do is we're going to grab a simple example which is going to be let's

imagine for a moment that we're interested in studying dreams right now I'm I'm fascinated by sleep I think

sleep is a really interesting topic I've never done any empirical work on sleep I can't say I've ever been terribly

interested in dreams but we're going to use this as an example uh just to just to chat a bit about how different

research methodologies could be brought to bear on the same sort of research question

so if what we were interested in was studying someone's dreams right one of the things we could do is we could use a

simple observational method right so we could simply try to observe people uh while they sleep for example now there

isn't going to be a lot we can tell about the actual content of their dreams from it from an outsider's perspective

watching someone sleep for example right so uh for example imagine if you have a roommate uh maybe uh tonight with their

permission hopefully right you go into their bedroom while they're asleep and you watch them while they sleep you

won't be able to get any real sense of the content of their dreams but one of the things that you could pick up on is

if you stare at their eyes you might be able to tell when they're actually dreaming because there's something that

we engage in called rapid eye movement basically our eyes start to move back and forth because the cornea at the

front of the eye sticks out a little bit right you can see what the eye moves back and forth as it pushes out the

eyelid a little bit and so what we could do through this process of observation is we can maybe

get a sense of when people are dreaming even if we can't necessarily get much insight into what they're dreaming about

now if we move away from observation if we move over to a survey and interview technique then we could actually ask

individuals about the content of their dreams so let's say for example uh that each day uh you when when your roommate

wakes up you sit down and you ask them about uh their dreams and so you go through this little standardized

interview that you ask them about every morning for the next month and let's say at the same time that you're doing that

with your roommate there are a number of other researchers that you're collaborating with who are also

interviewing other participants about their dreams and this would be a sort of interview

format right we could also get the same sort of content if we wanted by just asking people to complete surveys right

about their dreams so we can give people a little we could send people a link every morning via email or text and ask

them to fill out a survey about the content of their dreams so for example do you remember your dreams from last

night how many do you remember can you briefly describe the content of your dreams there were specific pieces of

information that we wanted like for example was your dream pleasurable did you enjoy the dream was it anxiety

provoking right we could ask those sorts of questions as well they're also standardized tests that are

used in many many areas of psychology so for example as I've mentioned previously I'm a social personality psychologist so

a lot of what I do is I measure people's personality traits so for example a lot of my work focuses on things like

narcissism people with kind of this grandiose sense of self right they have an over inflated sense of their own

abilities and accomplishments and those sorts of things they also tend to be entitled and so on so their standardized

tests within with tests within my area that have been developed by other people that I will often administer to

participants because they they are exceptionally useful for measuring particular characteristics like

narcissism or other personality traits right so standardized tester is quite commonly in psychology

now if for our current example of studying dreams there aren't a lot of standardized measures for uh for uh for

measuring uh different phenomena associated with drinks there are some but there aren't any that are

particularly popular right but we could Envision a world in which there was some sort of standardized test where if what

we were interested in was something like the anxiety that someone experienced in last night's dreaming for example right

we could imagine having a standardized test about dream anxiety or something of that sort

the next sort of research method is going to be something called a case study a case study is going to be a kind

of intense analysis of a single person's experience right so for example earlier I mentioned the idea of repeatedly

interviewing your roommate each morning about their dreams the previous night right and if you were only doing that

with your roommates right that would be a case study where what you're doing is you now have this kind of intensive

examination right of your roommate's dreaming either the content of their dreams or how often they remember their

dreams right whatever it is that you're asking about but that would be a case study if you're doing it with one person

right case studies can be uh can be helpful and informative especially in the early stages of trying to understand

some sort of phenomenon but typically what we want before we start really engaging in this kind of generalization

process is we usually want to go to move Beyond case studies usually case studies can kind of open the doors to helping us

think about things in different ways but typically we want larger groups of participants before we start really

engaging kind of the generalization that we're looking for in the research process

next up is going to be correlational research and so this is done when we don't actually manipulate any of the

variables at hand but rather what we may do is we may measure different things and see if they're connected to each

other so for example we could do something like we could measure someone's kind of

general anxiety or depression you know kind of mental health related phenomena and then we could measure the content of

their dreams like for example if an evident if an individual is feeling stressed and anxious because of their

work or their relationship or something else does that also have any relationship with the content of their

dreams right so for example the same person who's really stressed and anxious when they're awake may also report

having really stress and anxiety-laden dreams as well and so in that case we could look for

correlations between variables are there patterns of Association right as one thing goes up does the other thing tend

to change in a systematic Way Or Not Right But the key thing with correlational research is that nothing

is being manipulated right the things are just simply being measured the last type of research method that

we'll want to talk about now is going to be the idea of experimental research experimental research is going to be a

little bit different than correlational research right with correlational research again nothing was being

manipulated with experimental research we're going to manipulate one or more variables to

see what their consequences are for some other variable so for example if we stick with the

outcome variable right our dependent variable being something like the level of anxiety in someone's dream

right or in someone's dreaming State what we could do is we could try to manipulate right the something about the

person's sleep right so maybe for example we give someone we give we give we have two groups of participants an

experimental group in a control group we give the experimental group some sort of drug that's going to create really kind

of vivid dreaming experiences the control group we also give them a pill but it's a sugar pill right it's a

placebo that shouldn't have any real impact on their dreaming and then so we give both of these groups

the respective pills we then we have them in a sleep lab they sleep that night and the next morning we interview

them about the content of their dreams and what we could look to see is do the people in the experimental group who got

the drug that was supposed to make their dreams much more Vivid do they report different sorts of content or dreaming

experiences than people in the control group who simply had the placebo right and as we go on today we'll talk more

about experimental methodologies and correlational methodologies let's talk a bit more about an

experiment okay so what an experiment is doing is we're looking for cause and effect

relationships what makes something an experiment is that we're going to manipulate one variable for the moment

we're going to keep things we're going to start with the simplest type of experiment we're going to manipulate one

variable under control conditions so that any resulting changes in another variable can be observed right and again

what we're looking for are cause and effect relationships we're manipulating what we think is the causal force and

looking to see what its effect is on some other variable so two terms that you'll absolutely need

to know for quizzes and exams will be an independent variable and a dependent variable okay so the independent

variable which will often abbreviate as being an IV is going to be the variable that you actually manipulate right so if

we go back to some of our previous examples I have already used an example a couple of times of giving people who

are depressed either a real drug or a placebo that we think is going to have an impact on their level of depression

right in that example right that would be an experiment and our EX in that example the independent variable would

be the drug right so for some of the participants we're going to give them the actual drug that we think is going

to increase the level of Serotonin for example and others in the control condition won't get an active drug

rather they'll get a placebo write a pill that isn't actually going to have much of an active impact on them and so

what we're doing is we're manipulating right the dose of the drug right some participants are getting a dose of the

actual drug some are getting zero right of the actual drug because instead they're taking a sugar pill right so the

drug level right is what's being manipulated the dependent variable is going to be

the variable that's affected by the manipulation or at least the variable that we think will be affected by the

manipulation right here we're not manipulating the dependent variable right rather we're looking at the

consequences right of manipulation of the independent variable on the dependent variable right we're just

measuring or observing the dependent variable right so what we're interested in is how does X right affect y with X

being the independent variable y being the dependent variable right so again if we go back to our a depression example

from previously right the dose of the drug is our independent variable right we're manipulating the dose of the drug

people in the experimental group get the drug people in the control group don't get the drug right rather they get a

sugar pill instead what we're measuring for both groups is going to be their level of depression

right earlier we talked about measuring that with the self-report measure like the back depression inventory right so

we manipulate the level of the drug that participants receive and then we measure their level of depression through the

self-report measure and what we're interested in doing is comparing the two groups

to see is it the case that participants who got the dose of the drug are they reporting lower levels of depression

right than those who got the sugar pill or Placebo right if that's the case then we're seeing support for the idea that

the independent variable is having a consequence for the dependent variable right that X is affecting y

okay so just to make sure that we're all on the same page let's go through a couple of examples to make sure that you

can identify the independent variable and the dependent variable because these things would be fair game for quizzes

and exams okay so example one a researcher is interested in how heart rate and blood pressure are affected by

viewing a violent film sequence as opposed to a non-violent film sequence think for a moment about what the

independent variable would be in this situation and what the dependent variable would be

okay so hopefully right you were thinking about well the dependent variable right the outcome is going to

be the heart rate and blood pressure right that's your dependent variable that's the outcome right what is that

being impacted by well it's being impacted by the independent variable which is the type of film sequence that

people are exposed to right violent film versus non-violent film right so the film secret type of film sequence is the

independent variable heart rate and blood pressure are are the dependent variables

okay second example a social psychologist investigates the impact of group size on participants conformity in

response to group pressure so in this example what's the independent variable and what's the dependent variable

okay so hopefully you're able to identify that well okay our outcome is we're interested in what's going to be

impacting conformity in response to group pressure so that's going to be our dependent variable and the independent

variable for this example is going to be group size right so maybe for example the social psychologist has two groups

one group that only has three people in it and another group that has 30 people in it and they're interested in well

what do participants do if their group is small versus large right are they more or less responsive to group

pressure depending on the size of the group okay hopefully the difference between

independent and dependent variables makes sense to you okay let's go into a little more detail

about the idea of experimental groups and control groups the experimental group is going to

consist of the participants who receive some sort of special treatment right in regard to the independent variable right

so for example if we go back to our depression example from earlier right the experimental group be the people who

are getting the dose of the drug that we think is going to impact them right so for example getting the drug that's

actually going to increase the level of serotonin in their brain right that would be our experimental group

we're then going to compare them right in the simplest two group type of experiment with a control group the

control group should have participants who are really really similar right to those in the experimental group except

for the fact they shouldn't get the special treatment right that's given to the experimental group so for example

with our depression example we could have participants in the experimental group who are getting the dose of the

drug that's supposed to increase their level of Serotonin whereas the control group is instead getting the placebo

right to Sugar Pill right that shouldn't have any impact on the level of Serotonin

then after we've given either the real dose of the drug or the placebo to participants we then eventually measure

their level of depression and we compare the experimental group to the control group again if there's support for our

ideas right what we should see is that the experimental group should differ from the control group and in this case

what we should see if our ideas are correct is that the experimental group that's getting the dose of the drug that

should increase the level of Serotonin right should be reporting lower levels of depression right basically increasing

their serotonin levels right should make them feel less depressed right should help alleviate right their depressive

symptoms right relative to the people in the control group another thing that's going to be a

pivotal part of an experiment is going to be the idea of random assignment random assignment is when we have a

group of participants and we have a we have an equal chance right of any particular participant being assigned to

any group or condition in our study so let's say for example in our depression example we have 100

participants what we would want to engage in is random assignment where once we have our group of 100

participants right who have agreed to be in our study if we have two groups an experimental

group in a control group what we would want is to assign 50 of them to the experimental group and 50 to the control

group and we'd want to do some sort of random procedure to assign people right the reason why this is so important is

that if we're able to randomly assign participants to condition right any difference that we see between the

experimental group and the control group is much more likely to have been to be due to whatever it was that we

manipulated right the manipulation of the independent variable right now if we fail right in this random assignment

procedure let's say instead of randomly assigning participants we decide instead what we're going to do since we want

half of our participants in the experimental group and half in the control group

what we're going to do is we're going to have our 100 people who have all agreed to come in for this study but then

instead of randomly assigning them like we know we should we decide to be lazy and we're going to take the first 50

people and put them in the experimental group and the last 50 people who show up we're going to put them in the control

group right the problem with that is there could be some difference right between the people who show up first to

be in our study and the people who show up last right now instead of randomly assigning people now we have them we

have a pre-existing difference between our groups maybe there's some difference between the people who arrive early or

at least on time for a study compared to those who arrive later right those pre-existing differences now can make it

very muddy in terms of understanding well are people in the experimental group are they responding to the dose of

the drug they received or are people who arrive early right and are maybe more conscientious right do they just tend to

recover faster from depression right so that's why random assignment is so important what we want is to make sure

that the groups really are as equal as possible on almost any sort of variable we could think of right random

assignment especially as the groups get larger and larger will help make sure those groups are in most cases going to

be relatively equal on most characteristics we then are going to manipulate the

independent variable for only one of the group so for example in our simple two group experiment we're going to have a

control group and an experimental group and the experimental group is the one that gets the active dose of the drug in

our depression example and again the resulting differences that we see when we followed this basic

procedure it's going to be most likely due to the independent variable right now we haven't completely and utterly

ruled out other possible explanations for example it's still possible right that it could just be a fluke right that

our two groups differ right things happen right now as the sample sizes get larger and larger right becomes a

smaller and smaller probability event that a large difference we see between our groups really is due to random

chance right but we can't completely and utterly rule that out right so what we're talking about here are still

issues of probabilities right as group sizes get larger and larger and as the differences between groups get larger

and larger it becomes increasingly likely right that what we're seeing is a difference it really is due to the

influence of the independent variable we also want to talk about extraneous variables extraneous variables are going

to be any variables other than the independent variable that seem likely to influence the dependent variable in a

particular study so for example with our depression example that we were talking about previously there are a number of

other extraneous variables that may also impact someone's level of depression so just to give a one simple example

gender tends to be associated with depression in general women tend to report higher

levels of depression than men do and so gender in our depression example could be considered to be an extraneous

variable because it's another variable other than the independent variable that might actually exert some influence over

the dependent variable in a particular study a confounding of variables occurs when

two variables are linked together in a way that makes it really difficult to sort out their specific effects right so

now following up with our depression example let's imagine we made the really awful choice of confounding right our

independent the levels of our independent variable right the dose of the drug either getting the active drug

or the placebo with the gender of the participant right so let's say for example we gave the we gave the actual

dose of the drug right to men right and for the placebo we gave that to women right so now what we have is a perfect

confound which is a really really big problem in research so let's say that we give the experimental dose of the drug

to our male participants we give uh the placebo to the female participants now let's imagine we conduct our study so

the men are taking the actual dose the women are taking the placebo after a period of time goes by we

measure depression and sure enough we find the people in our experimental group reporting lower levels of

depression than people in the control group however now because we have this confound we have no way of separating

out whether it's really due to the dose of the drug which is what we were intending to have a consequence for

depression or is it due to the fact that we gave men the active dose of the drug in women the placebo right it's possible

right that the difference in depression could be due to the dose of the drug like we expected but because of our poor

design of the study it's also possible that it could be due to the gender of the participant right maybe it's because

men report lower levels of Oppression than women maybe that's why our experimental group is reporting lower

levels of depression than the control group not because of the dose of the drug but because of the gender

composition of those groups so this is why it's really important to avoid confounds right what we want to do

in experiments is to make sure that our groups are as equal as possible Right on everything other than the variable we're

manipulating right so we don't want there to be confounds other variables that are so closely linked with our

independent variable that we can't separate out their their effects so let's go through a basic experiment

just to make sure again we're all on the same page so we start off with a hypothesis in this example so our

hypothesis that starts to study is that we believe that when people are anxious they show a desire to affiliate with

others so they want to be closer to other people so what we would do next now that we

have our hypothesis and we have our participants now we're going to randomly assign them so we're going to assign

subjects randomly to an experimental group and a control group right so let's say again we have a hundred people just

to keep our numbers nice and round so we have a hundred people we're going to randomly assign them through some

procedure to our experimental group or a coin or a control group now this can be done through something as simple as just

kind of flipping a coin for each person heads you're in the experimental group Tales you're in the control group

they're more sophisticated sorts of procedures you can use but what you want is something that's random where it

isn't uh their assignment to the experimental group in the control group isn't based on the first people to come

into the lab get excited assigned to the experimental group the last people into the control group or some sort of

pre-existing demographic feature like all the males go in the experimental group all the women go in the control

group because that's going to create confounds and problems where you're research

we then manipulate the independent variable so in the example we're talking about which is drawn from an actual

study the experimental group would they were told that they were going to be participating in a study that would

involve them receiving some really painful electric shocks um and what what happened for the people

in the experimental group is they're told okay well you're going to get these really painful electric shocks in a few

minutes but the equipment isn't quite ready for you so we need you to wait for a few minutes and by the way while

you're waiting you can either wait in this room by yourself or you can wait in this other room with some other

participants who are also waiting to participate in the study people in the control group were given

essentially the same information of what was changed for them was that the shocks were they were told the shocks would be

mild and painless right so they were told you're going to be participating in a study you're going to get some

electrical shocks but they're going to be relatively mild and painless oh by the way the equipment isn't quite ready

for you it's going to be a few minutes so while you're waiting do you want to wait in this room by yourself or would

you rather wait in this room with other participants who are waiting for this to participate in the study

and the measurement of the outcome variable the dependent variable would be which group which room do people choose

to wait in do they choose to wait in a room by themselves or do they choose to wait in the group with other people

right and so if the results show the people in the high anxiety group right Illustrated a greater desire right to

wait with other people than people in the low anxiety group did right this would lead to the conclusion that maybe

anxiety really does increase our desire to affiliate with others at least as measured in this particular study right

now again we have issues of operationalization right we use an operational definition of what do we

mean right by anxiety right well here anxiety we created right by either assigning people to a condition where

they were told they were going to be receiving really painful electric shocks or we created a low anxiety condition

where they were told you're going to be receiving some shocks that are going to be relatively mild and painless right we

also had an operational definition of our outcome period right this desire to affiliate which was uh captured based on

whether or not they said they wanted to wait in a room by themselves or they would prefer to wait in a room with

other people right that's how we measure the desire to affiliate there are also some other variations on

experimental designs I just want to mention there there are a whole host of variations I just want to mention a few

of them uh here for our purposes one is that what we've been talking about previously is something called between

subjects design where different participants are taking place in the experimental condition relative to the

control condition however we could use a variation of this where the same group of individuals

right is exposed to two different conditions so for example with our high and low anxiety condition we talked

about a moment ago we could have something where we have participants come back on two separate weeks right

and one week they're exposed to the high anxiety condition the next week they're exposed to the low anxiety condition we

kind of alternate that between participants but that would be a within subject's design where instead of having

different participants for the two conditions everyone is experiencing both conditions and that's within subjects

design one of the nice things about within subjects designs is a reduce extraneous

variables right now the same participants who are in the high anxiety condition are the same people in the low

anxiety conditions so we don't have to worry about any demographic differences for example that may happen to uh to

emerge between our two uh two groups of participants we can also use more than one dependent

variable right this may be helpful in terms of compiling a more complete nuanced picture of the impact of the

independent variable right so for example in our high anxiety and low anxiety experiment we were just talking

about we looked at the desire to affiliate right as being our primary dependent variable but there could be

other things that we could look at as well right we could measure right things like their uh their affect right is the

person happy are they sad are they anxious are they angry you know so we could look at other dependent variables

besides just the desire to affiliate right not saying there's anything wrong with the desire to affiliate in that

study but we could look at other things in conjunction with the desire to affiliate right to get a a bigger

broader picture of the impact of the independent variable we can also manipulate more than one

independent variable at a time what this allows for is for us to study interactions between variables which

will be very important in trying to capture the complexity of the world around us so let's talk about that in

more detail so let's imagine we're going to manipulate let's imagine what we're

interested in is measuring test performance and so we're going to manipulate two

different independent variables simultaneously okay the first independent variable that

we're interested in is the presence or absence of distracting music right so distracting music is up here so for half

of our participants we're going to have distracting music being present during an exam

for the other half the participants the distracting music is absent right so half get the distracting music half

don't at the same time our second independent variable is going to be the temperature

of the room for half of our participants the room is a normal temperature let's say 75

degrees or something of that sort for the other half the participants we make the room uncomfortably hot let's

say we make it 90 degrees right so now what we have are these four different cells of our design right for a quarter

of our participants they're taking the test in a room that has a normal temperature but there's distracting

music that's present for a quarter of the participants that are taking the test in a room with

normal temperature and there isn't any sort of distracting music for a quarter of our participants are

taking the test in room that's both hot right and also has distracting music played

and for the final quarter of our participants are taking the test in a room that's hot but doesn't have any

sort of distracting music if what we're thinking is going on is that both the presence of new distracting music and a

room that's uncomfortably hot May both impair our ability to perform well on tests if what we would expect to see

then is that this group here may perform the worst right these folks here may do the best right and these two groups that

have only one of the two types of unpleasant environmental conditions may do somewhere in the middle

so some of the strengths and weaknesses of experimental research so one of the really really vital strengths is it

allows us to draw cause and effect conclusions from our studies couple weaknesses though with

experiments is that there's an artificial nature to experiments in many cases so for example uh if we talk about

uh things like our our high and low anxiety example if you come in for a study and the experimenter tells you oh

you're going to get some really painful electric shocks right since you're participating in a psychology study you

may be suspicious about what the nature of the study really is right and so there's an artificiality that that kind

of permeates a lot of psychology studies that may impact our Behavior right so for example you may be suspicious and

skeptical about whether or not you're really going to be subjected to these really painful electric shocks or not

right and as a result your behavior may change because of the fact that you first of all know that it's an

artificial situation and you know that you're being observed right you know that you're participating in a

psychology study there are also some ethical and practical issues with regard to

experiments there are some independent variables that we simply cannot manipulate for ethical or practical

reasons for example psychologists are often quite interested in the long-term consequences of things like childhood uh

traumas and negative experiences like childhood abuse however for both ethical and practical reasons we can't

manipulate those things we can't subject some children to a early life experiences of abuse and others we

protect from that right that's not the way the world works for both ethical and practical reasons rather when we're

interested in the long-term consequences of these traumatic early life experiences rather we we can't do

experiments in many cases rather what we have to do is have to rely on correlational Research where we're

simply measuring these things that have or haven't occurred and seen what the consequences are right without being

able to manipulate those experiences now let's turn our attention to a complementary type of research method

which is a correlational methodologies so correlational methodologies are used when a researcher can't or ethically

either they practically can't or they ethically can't manipulate the variables being studied

um so there are a number of different types of correlational methodologies that are used uh one is naturalistic

observation this is when a researcher engages in careful observations of behavior without directly intervening

with the participants so for example if we were interested in aggressive behavior exhibited by

children one of the things we might do is we might go out to an elementary school with permission of the school of

course and we may observe kids out playing on the playground during recess and so we might observe children playing

without making any sort of direct interaction with the kids maybe we just have our clipboard or tablet and we're

simply recording whether or not kids are biting or punching or kicking other kids during a certain period of time right so

that would be the idea of naturalistic observation we're simply recording what is going on without actually intervening

in any way um second type of correlational methodology is a case study a case study

is going to be an in-depth investigation of an individual participant so for example we'll talk about Freud in a

number of places over the course of the semester what Sigmund Freud did is he performed case studies he would talk

about patients that he was seeing and he would he would write about their experiences and the therapeutic process

he was engaging in with that particular patient right and that's a case study again case studies are very useful but

typically we want to go beyond that in the research process we can also use survey research this is

using questionnaires or interviews interviews to gather information about specific aspects of a participants

background or behavior as I mentioned multiple times already I'm a social personality psychologist a

lot of the work that I do is of this type of survey research nature so for example if I'm interested in the

connections that narcissism has with certain types of social behaviors I can't manipulate someone's level of

narcissism very easily at all right I don't have that ability to do I can't do that very very easily or very well

so typically a lot of my work involves measuring their level of narcissism and then maybe measuring some other variable

that I think narcissism is going to be connected to so a lot of times I rely on survey research not exclusively I also

do experimental methodologies as well but a lot of my work is done using survey research

if you're engaging in any of these sorts of correlational methods what you're typically going to be looking at will be

correlations between variables that are involved so one of the things you'll be responsible for knowing for quizzes and

exams will be what are correlations and what do they mean and what are positive versus negative correlations so let's

spend a bit of time talking about that so what a correlation refers to is a relationship between two variables

meaning if there's some sort of systematic change between the two variables let's start with a positive

correlation a positive correlation tells us that variable x a high school GPA for example and variable y college gpa are

going to change in the same direction right so if two things are positively correlated as one thing tends to go up

the other thing is also going to go up right so across people right as they were reporting higher and higher levels

of high school GPA they also tend to report higher levels of college GPA now what's also a positive correlation

is if the both of our variables X and Y are moving in the same direction downward right so as individuals were

reporting lower and lower levels of GPA in high school their college GPA is also lower and lower right the idea with a

positive correlation is that X and Y are moving in the same direction they're both going up or they're both going down

in contrast a negative correlation is X and Y are moving in opposite directions right so as X goes up y goes down so for

example as the number of absences that people are reporting as those go up in a class then their exam scores are likely

to go down right they're moving in opposing directions right that's the idea of a negative correlation right as

absences are decreasing right exam scores are going up right so negative correlation X and Y are moving in

opposite directions for a positive correlation they're moving in the same direction

some of the strengths of correlational methodologies is they allow researchers to describe patterns of behavior and

discover links or associations between variables that we can't necessarily look at using experimental methodologies

right so one of the things this is going to use it's going to broaden the scope of the phenomena that psychologists are

able to study right for example we mentioned earlier the idea that psychologists are really interested in

the long-term consequences of things like childhood abuse right but we can't manipulate that right we can't

manipulate those sorts of abuse experiences so rather what we can do is we can measure those abuse experiences

what what do people say they experience during their childhood right or what what do official court records or police

records tell us they experienced and we can then see what that's associated with later in life

one of the huge weaknesses right of these sorts of correlational methods is they can't be used to imply causation

right so correlation does not imply causation just because two things are connected because they tend to move in

concert with each other or move in opposition to each other right it doesn't mean that one of those things

causes the other right and this will be an issue we'll return to very soon in an upcoming lecture