Aims and Hypotheses

Hello. My name is Owen O'Sullivan.

I'm an associate lecturer at the University of ST Andrews.

And today, in this lecture, I'm going to talk about themes and hypotheses.

So to start going to remind you that psychology is

a science and a scientific topic follows the scientific process.

So that process is kind of circular.

We observe something in the world, and we might have a question about it.

A question about how people behave or why people do certain things.

Um,

so once we make that observation that we want to study as we want to learn more about it

And so we develop an aim we want. Our aim is to understand that thing more.

We use hypotheses, uh,

to make specific predictions that help us test our questions.

We go out into the world, we collect some data.

Once we've collected our data, we analyse it and interpret it.

And this allows us to develop theories or richer

understandings of phenomena and the ways that people think,

and the ways that people behave.

And so, in this circular research process,

we now are going to talk about aims and hypotheses.

So when beginning a study, you need to have a kind of broad aim.

What is the main thing you want to achieve from a piece of research?

So an aim is this overarching goal of the project that you are starting to design.

It will often be written or presented in terms of the topic

that you're interested in or the theory that you're attempting to test.

So the best way of explaining a name is to think of an example.

So I'm going to use an example from the history of psychological research.

I'm going to talk about Albert Pandora and his social learning experiments.

Bandara conducted a range of experiments in around the

19 sixties to understand how Children learned certain behaviours.

In particular, he was interested in how Children learned aggressive behaviours.

Now, around this time in psychology,

psychologists were very interested in how people learned.

But there was a focus on individual learning

how we learned through our own experience.

But Pandora observed that that's not how humans learn all the time.

We also learn from other people's experiences

by observing how other people behave,

and so so in his experiments,

Albert Pandora had an aim to understand

how Children learned socially about aggressive behaviours.

So here you can see the aim is quite general.

It is overarching,

and through that aim he can conduct a number of different research

studies to better understand how Children

learn aggressive behaviours from others.

So in his experiments he designed some very

neat little studies where he recruited Children.

And he, um, he got them to observe the behaviour of adults interacting with a toy.

That toy was famously called the Bobo doll.

And so he observed,

adults or the Children observed adults interacting with the Bobo doll.

And some Children observed the adults interacting

with the doll in an aggressive manner,

kicking and punching.

Or they're different Children,

observed adults interacting with the Bobo doll in a neutral manner.

Uh, not in any way, uh, interacting with adult in an aggressive way

and so on. And so, through this method, Bandara wanted to understand

whether that would have an impact on the Children's behaviour,

and the Children would interact with that dull at a later time point.

And so with that experimental setup, Bandara developed a hypothesis.

So too,

so a hypothesis is a clear and specific statement about what you predict

from your research design from your theory and that you have developed beforehand.

It will help you achieve your aim.

It will help you to better understand and that particular idea or theory,

and in comparison to a name, it is a lot more specific.

It is presented in the context of the things

that you are changing and manipulating in your study

and the things that you're looking at in terms of an outcome.

More specifically,

is often phrased in the context

of independent variables and the dependent variables

thinking about the the Pandora example, the social learning experiment.

What Pandora is manipulating is the experience of

the Children before they interact with the doll.

What he changes is whether the Children see an adult interacting with

the doll in an aggressive manner or in a non aggressive manner.

So that's the independent variable what they socially observe.

Later on, he will measure, or he measured

the Children the Children's behaviour when they were interacting with the dull,

and he specifically measured the amount of aggressive

behaviours that the child performed on the doll.

So that was the dependent variable.

And so from this research design and

when thinking about these specific behaviours,

Bandara proposed a specific prediction hypothesis.

This prediction was

that

we would expect that Children that observed the

aggressive adult models will behave in a more

aggressive manner towards the dolls in comparison to

the Children that didn't observe the aggressive adults.

So here we have a very specific prediction couched

in terms of the independent and dependent variables.

Now here I have presented something that's called an experimental hypothesis.

Another term for this is the alternative hypothesis.

This hypothesis predicts that there will

be a relationship between your variables,

that there will be an effect something will be going on,

that there will be a difference between the two conditions in this case,

the Children that saw the aggressive model and the Children that saw the neutral,

the aggressive behaviour

and the behaviour that was more neutral.

But alongside this alternative hypothesis, we have a null hypothesis,

and this hypothesis states that there

will be no relationship between the variables

that there'll be no difference between the two groups that you have changed.

And now you may ask,

why do we use the null hypothesis when you

can kind of infer it from the alternative hypothesis

and the reason we use the null hypothesis is because the maths that we use the

statistics we use to test our hypotheses are

based upon this concept of the non hypothesis.

And so you may learn about null hypothesis testing at a later stage.

But for now,

it's just important to know that the null

hypothesis is that specific prediction that there is

nothing going on in your study that there

is no relationship between your two variables.

While the alternative hypothesis

is the statement that there is a relationship between your two variables,

another thing we need to think about when we've got when we're

presenting our hypothesis is whether it is one tailed or two tailed,

UH, two tailed hypothesis is also called a non directional hypothesis.

It states that if there is a relationship, it will occur in a particular way.

So in the example I discussed about Pandora and we could predict

that there will be a difference in aggressive behaviour acted towards the dull

based upon whether the Children observed the

aggressive interaction or the non aggressive interaction,

and we could make that prediction without assuming that

one group would be more or less aggressive.

That's a two tailed hypothesis.

You're making a prediction that there would be a difference between your groups,

but you're not stating in which way that difference will exist.

However,

the difference that I did or the hypothesis I

did present earlier was a one tailed hypothesis.

We predicted that Children that observed the adults

behaving in an aggressive way that those Children would

themselves behave in a more aggressive manner towards

the doll in comparison to the other Children.

So there is a direction in that prediction.

We expect one group to perform more of a

particular behaviour or perform differently to the other group.

Now it's important that these one tailed hypotheses

are based on theory or previous observations.

We can't just pluck them out of the air,

but they are important because they do inform the types of

analyses that we will conduct after we've collected our data.

So I hope that they're in those examples.

You've got a clear idea of what an aim of a research project is

and the different types of hypotheses we

used to make predictions in our research studies