Methods - Experiments

Hi and welcome to today's lesson.

We are looking at methods now, so it's a whole

new topic on today.

We're going to start by learning about met.

The method of experiments on this is in order to

understand the key information about experiments so we know the

basics off them on.

This leads towards us understanding methods in sociology because this

is a new topic.

So over time we look at each of the different

types of methods.

Today, our focus is on experiments on to start with

a bit of a definition, so experiments are all about

using what's known as the scientific method.

The point here is that we're trying to control the

variables, the things that we can control, things that might

change in an experiment so we can determine a clear

cause and effect.

Now, if you were to take an example of an

experiment using the scientific method, perhaps you look at growing

plants so you might get a section of the same

point. You might get a whole bunch of the same

type of plant and you would set up their conditions

to be exactly the same.

You're getting the same amount of water the same amount

of light given the same soil, the same conditions.

Exactly. And then, for the experiment, you choose one variable

to change, to see the effect.

So perhaps on this has happened to you would play

different types of music to the plants.

So what you would happened, you'd have you determined one

variable. One thing that's different between those different groups on

dhe. Then you could see the effect.

You could see the cause and effect.

What type of music has an effect on part, if

any type of music.

But to make sure that experiment works, using the scientific

method, you need to control the variables.

No, in that plant example parables with soil, the sunlight,

the water, the temperature, all the other things that we

could control.

Those things that might be different so that we can

rule those out is why the change happened when we

make sure that there's only one thing that's different on

when sociologists undertake experiments, that's what they're trying to do.

They're trying to control as many of those variables as

they possibly can, so that they could determine one thing.

The causes that leads to infect one cause that leads

to an effect.

No. Um, so we're talking to here the border experiments

specifically, That's the first time we'll look at some other

top two experiments.

But this is where the sociologist has the most control

over the variables.

There is a question whether this is possible with human

beings will talk about that later.

Is it possible to actually control the variables involved in

a human being?

Or are we far too complex for experiments?

Toe have those controls.

But the what they're looking at here is the drug

variables. This means it's a very reliable type of X,

that of sociological sociological method, because we can repeat it

exactly the same.

We can set up that laboratory exactly saying each person

comes into that oratory.

We can keep things the same for them, apart from

the one variable that want to change the one thing

we want to change so we can control everything in

that situation.

And that's what's important with this type of method, specifically

laboratory experiments, because we can control that lap.

It's not in the real world.

There's no all the things that come along with the

real world there that might affect people we control as

much as possible.

Now, when we talk about methods, we often break down

on our social.

Just the positive ists and interpretive ists now are positive

ists. They want to control the variables.

They want to treat sociology like a traditional science.

They want to treat it like physics or chemistry or

any of the other traditional sciences.

They want to be able to control the variables, and

they won't really reliable, repeatable representative studies.

They want to study as many people as possible on

be able to repeat that in exactly the same way

so that they can draw cause and effect positive, etc.

Sociology. Like other sciences, they see it like a traditional

science. But we also have interpretive.

It's sociologists now interpret.

This is a far more concerned about the natural behavior

people exhibit.

They're concerned about undertaking methods that see people in their

natural way, see people acting in the way that they

would if they didn't know they were being looked at,

that they were being studied.

They're interested me interpreting people's behavior.

So interpretive, sociology, social just are not keen on the

bar trick.

Perhaps they're not keen on experiments in general, especially liberal

friends. They see them as a ZX artificial situations.

They don't think that they're very valid.

They don't think the very true representations of people's behavior

is not valid representation of their behavior.

Where is the positive ist who are far more concerned

about representative on reliable nature, the repeatedly it repeatability of

an experiment or a method They like the larger experiment.

Another experiment, because it can be repeated.

You can control that.

You know what your variables are.

You know what you're controlling, The change it so you

can repeat that again and again with different people in

the exact same situation.

So if we think about this, then there's a criticism

we spoke about.

Society may be seen as far too complex to actually

control the variables.

Perhaps human beings are wet and society itself away to

complete. There's too many things going on, so therefore we

can't control those variables.

We don't know all the stuff that's going on with

the human being.

We might get that human being into the lap and

control the direct environment that Aaron.

But there's all of the things that happened to them

that day that week that year in their entire life,

all of those things that have gone into creating that

person's personality.

Yes, you may be on to break it down by

looking up men and women or people from different, um,

ethnic backgrounds or people with different sexualities and then on

bond, make judgments about groups like that, like possibly should

like to dio.

But in tough it tough to this would argue that

no, the people's the input to those people throughout their

life is far too complicated.

Humans of Barton complicated to narrow down the variables.

Therefore, it's no very valid data is not true data

about how those people would act or what they think

doesn't give us a true understanding of society.

Another problem with barks experiments is that they can only

be used for because small sample, so you can only

have a small number of people doing them.

One of this is because it would be quite expensive

to rent Barker's space set up, but also if you're

running experiments and have taught people coming in taking part

in that if you think about the society and the

number of people in that society, If you're under the

bark experiment, you are only evident in a very small

sample. From that only a small number of people You

also have to question which people are gonna be involved

in this.

People might have the time or inclination to go into

a laboratory and take part in some kind of experiment.

So again, that's going to skew the data.

That's gonna give us a slightly different group of people

who are involved.

So for and talk to this again they started Robot

experiment is not a good form of sociology because of

all these problems, because it doesn't give us a valid

understanding of people.

Where is the positive ist sociologist Stacy?

It is a very important way of undertaking sociology, or

many of them do because of that controlling the very

but variables on because of that reliability, that fact that

we can repeat it again and again and again.

Some other problems we have with the barter experiments with

undertaking this kind of sociology, controlling those variables with setting

up something in a lab of bringing people in in

the ethical Brant's ethical bombs means within corrupt the morals,

the right and wrong what what's right and wrong.

And actually, is it is it right to experimental people

is okay to put people in this situation on run

experiments involving them.

Now, we might not be talking about expense that directly

caused physical harm to somebody, but there could be some

kind of emotional harm involved.

And when we look at some of the examples of

this, you see, that links examples at the end, and

one of the lessons later in this week is to

look at one of those examples.

But there could be emotional harm caused by this type

of method by its setting up a lap Ah, nde

setting an experiment whereby you, um, make people act in

a certain way or you do something somebody to see

how they react can also it can't involve people knowing

what's really happening.

Most of arch experiments keep the rial idea a secret.

Now, some some don't some kind of let people involved

in what's happening.

But the majority of them actually don't tell the person

what's being experienced, because by telling them you're gonna affect

their behavior on, they might try and act in the

in the way that they think is right to act.

Because of that, that's gonna ruin the data.

So it's not seen as ethically right either anymore to

not keep people what's called informed consent to let them

know exactly what they're taking part in.

Why so Elasticsearch generally doesn't get people informed consent.

You don't really tell them what you're looking for.

Offer on dhe again.

Just the the ethical problem of running an experiment.

Putting people in a situation that may cause them some

form of harm, not necessary.

Physical remember, could be emotional, other kind of mental harm.

Distress things like that.

But there's also this problem of what's called the Hawthorne

effect. Okay, the whole effect actually comes from what's called

a field experiment, and I'll give you an example that

later I'll explain the experiment it came from exactly.

But what the Hawthorne effect means is that people will

act differently when they know that they're being observed when

they know they're part of some kind of experiment.

People would act differently if we were back in school.

If we were back in an actual classroom, imagine that

on. We had somebody come in tow, watch what was

going on in our lesson, then generally, people in that

classroom would act differently.

They may not even realize they're doing.

They would act differently.

If we have somebody coming in saying they were looking

for something specific, then we may act differently because we

may want to show them that specific thing, all the

opposite we made to ruin their experiment.

For some reason, we may not even know why.

But if we have somebody coming into our classrooms, and

I'm observing this fast to see how how boys act

in a classroom and to see what their behaviors like

and how much effort they put it.

Actually, some of the boys in that classroom makes suddenly

want to put in a lot more effort than they

normally do, because they may feel when I need to

show this person, I need to show them how I

actually showed my best side.

So assumes were observed, changes our behavior and therefore things

like lab experiments and other forms of expected experiments.

The very nature that something is involved in an experiment

and they know they're being watched will change their behavior.

And I'll tell you at the end, when I talked

to the examples where horse on that comes from Come.

But where is the love experiments then?

So is one.

Is that setting up a laboratory, having people in that

laboratory, setting up some kind of situation for them and

then watching what happens and writing down the results?

We also have what's called a field experiment.

Now field experiment takes place in the natural environment, so

it takes place in in people in their normal place.

That they would be could be a workplace could be

a school that, via Felix Salmon, could be someone's workplace.

That's why the whole phone comes from so you could

go to that place and then you could try and

control the variables in that workplace foreign, that natural environment.

So you're trying to get rid of artificiality of the

lab but still control the variables.

So it's kind of a compromise between, say, Yeah, I

get the lab is to artificial constantly.

Something's in the lab because it doesn't represent real society.

But we still want to control those variables, and I'll

talk about a couple of examples at the end that

might help you understand that the third and final time

is the comparative method on this is what's often called

a thought experiment.

So this doesn't happen in real life.

You don't actually go out and observe these.

What you do is you collect.

You have data on this could be used to study

things that happened in the past.

And this is the only type of experiment that can

study past behaviors and past actions.

But you identify groups.

You try and find groups where everything is the same,

whereas many of the variables are the same as possible

except for the one you want to study on.

Then you look at the data on those groups.

Then you look at the effect of Oh, so you

are not really doing the experiment.

You're doing all in your brain by looking at analyzing

data on again.

Hopefully an example will help you understand that.

So let's do this.

Let's talk about some of the examples will try, get

threw. These nice and quickly.

Some of the examples toe help you understand what these

experiments might look like.

So some relevance experiments.

We've got Stanley Milgram's experiment around control and wine.

People take actions, and later that week I'm gonna ask

you to do some reading.

What, you have a documentary about that.

What he did is he got some actors on DSA,

members of the public.

But the actors were pretending to be members of the

public, and he hooked them up to a machine.

One person had to ask the other person questions.

They couldn't see each other.

The other person had to answer those questions is them?

They remember them every time the person answered the question

about it.

Wrong person, asking the question, was asked to give him

an electric shock rising in in how severe the electric

shock loss.

Now, nobody actually received an electric shock.

The people who are answering the questions where actors pretending

to get an electric shock.

But what he wanted to look at is how far

people go on.

The electric shocks got to a point where they would

have killed somebody if they're if they're given that electric

shock on loads of people just went through with it

gave the electric shock.

They gave that person the electric shock, even though it

was it was such irate because somebody was telling them

to on that level of control.

So he set that up.

That was a Labat from.

He set up a Pretend machines electric shocks.

Another example is in Bardo Stanford prison experiment.

He was a professor at Stanford University on, he said

to the basement, to be like a prison.

He got students toe how many paid them.

Some of them were made as prisoners.

Some of them made his prison guards, and he watched

to see what dynamic would happen and how people treat

him. Basically, the prison guards started abusing prisoners.

Being really horrible to them treated really badly, degrading them

things like this.

And it was a completely false environment.

It was a fake president, my lab set up to

look like a prison, and he watched to see what

actions happened.

It's another really interesting one again.

I think that with the reading with a documentary to

watch, if you'd like to choose that one on, then

there's a dangerous experiment around aggression and bobo dollars, where

Children have gotten toe watch somebody treating this big giant

inflatable dog nicely or to see them treating it badly

and beating up.

And then he put the Children in the room with

the door to see what they did on the Children

who walks violence, acted on the door, acted out violence

on the dole themselves.

Andi argued that seeing violence creates violence so you could

see its a Slavic some he set something up.

He controlled all of the variables to control what those

Children saw to then decide what could happen to them.

So those are some examples of experiments.

So example field experiments.

So we talked about Hawthorne earlier, was actually a power

plant and electricity station, creating electricity on what this Elton

Mayo did is he went in, he said, I'm going

to study working conditions on how people work.

And he started changing loads of variables.

You talked to change anything.

So how light itwas how dark it waas, how hot

or cold?

Waas. What I found is when he made it nice.

Only had people come for chairs and get them better

lighting and warmer environment.

They worked harder, but where he also found that if

you made it worse, it made the lighting words and

made it more uncomfortable in medical situations.

Worse, they still worked harder on what he what he

deduce from this is if people are being watched, then

they work harder.

When we don't know that they're being observed, they work

harder. We've got Rosenstein jackets on teacher labeling where they

told teachers these kids have right.

These kids aren't right and then let them teach seven.

See what happens on.

We've got Rosen hands saying incent places where you took

perfectly sane people talking toe commit himself into a mental

hospital by saying certain things and see what it was

like in there.

I'm in the final works.

I'm running out of time on this video on your

on 50 minutes for these is the comparative type.

So this was dark.

I'm into data on suicide between Catholics and Protestants, all

the features with the same people apart from their religion.

But he found the Catholics were less likely to commit

suicide and argued this was because they have more of

a sense of community.

I've had to rush a little bit there in the

end, so make sure you do the reading as well

to back this up.

So you understand it.

Thank you very much.

Watching this on experiments.

That was a long walk by

Methods - Experiments

Audio TranscriptAuto-generated