Research methods

Scientific experiments involve testing…

cause and effect relationships

To test cause and effect relationships psychologist use the …

scientific process

5 steps of the scientific process

1. Aim

2. Hypothesis

3. Method

4. Results

5. Conclusion

Aim:

a statement which gives the reason or purpose for carrying out an experiment

Hypothesis

a prediction about what you expect to happen in an experiment

Independent variable

what we change in an experiment

Dependent variable

What we measure in an experiment

Level of Independent variable :

the number of different versions of the IV there are in an experiment.

Operationalising variable means…

defining how the variable is going to be measured

Results

changes observed during the experiment

Conclusions

interpretation of the results

If hypothesis testing leads to results that don’t support the hypothesis, researchers need to…

adjust their hypothesis.

Scientist use _______ to test hypothesis

the scientific method

Empircal evidence

information that’s collected through direct observation

Falsifiable

when a hypothesis can be proven false

Replicability

repeating a study to re- test the hypothesis

Control group

The level of IV that is not being manipulated by the researcher . This is used to compare with the experimental group.

extraneous variable

a variable other than the IV which affects the DV

objectivity

when a researchers personal bias does not affect the results of the study

A theory is…

A comprehensive explanation that has already been backed up by the scientific method.

In order to test predictions using our theory, our theory must be…

Falsifiable.

For a hypothesis to become a theory, it must…

have multiple experiments to back the results

Validity

when an experiment measures what it claims to measure

Lacks validity

when an experiment doesn’t measure what it claims to measure

The two ways we’ve seen that a study can lack validity are...

1. that the method of measurement doesn’t measure what it claims to measure.

2. that there are extraneous variables that affect the measurement.

When assessing the face validity of a test, we...

look at the test to see if it appears to measure what it claims to measure.

When assessing the concurrent validity of a test, we...

compare the results obtained to a previous test of the same behaviour, that is known to be valid.

We say that studies are reliable when…

the results of the study are consistent every time it is repeated.

A study can lack reliability if…

There are extraneous variables affecting the measurements. The test or questionnaire used to measure the DV is itself unreliable.

Internal reliability

internal reliability is when the individual items within a test are consistent with each other.

External reliability

external reliability is when the results of a test are consistent every time it is used.

The split-half method tests…

internal reliability

The split-half method

splitting the test items in two and comparing the results.

The test-retest method tests the…

external reliability.

The test-retest method

getting the same participants to take the same questionnaire or test on two separate occasions and comparing the results.

Three types of extraneous variables:

Participant variables, Investigator effects and situational variables

What are participant variables?

Personal characteristics of the participant which may affect the DV.

There are investigator effects in a study when…

The researcher influences the behaviour of the participants, and therefore the results of the study.

Situational variables are…

Extraneous variables that are features of the external environment.

Uncontrolled extraneous variables reduce both the

validity and reliability

Standardisation is..

a way of controlling extraneous variables.

How do researchers use standardisation to control extraneous variables?

Researchers make an extraneous variable the same for all participants, so that validity and reliability increase.

We can use standardisation to control…

investigator effects, situational variables.

We can’t use standardisation to control participant variables because…

we can’t change participants’ personal characteristics.

Matching means…

making sure a particular characteristic of the participants is divided equally across groups.

Random allocation is when…

Participants are assigned at random to experimental groups in order to ensure similar participants in each group.

definition of a single blind study

The participant doesn’t know which experimental group they are in.

What is meant by demand characteristics?

Demand characteristics are aspects of the study which lead participants to guess the aim of the study and form expectations about how they should behave.

What is meant by standardisation?

Standardisation is a way to control for extraneous variables, by making the procedures and extraneous variables the same for all participants.

What is meant by matching?

Matching means pairing participants across different groups so they’re similar on a specific characteristic (for example, age or ability), so that characteristic is evenly spread across the groups.

What is meant by random allocation?

method of chance to control bias during sample selection

What is one limitation of matching or a matched pair design?

A limitation of matching is that we cannot control for every possible participant variable that could affect the results of the study.

Internal validity

A study has internal validity when the study measures what it claims to measure, and there are no extraneous variables.

External validity

A study has external validity when the results of the study generalise to other situations and people.

Ecological validity is when…

The results of a study generalise to how people behave in everyday life.

When does a study have temporal validity?

When the results generalise across time.

A sample is…

the target group of the researcher used in the experiment

Types of sampling :

1. Volunteer sampling

2. Opportunity sampling

3. Systematic sampling

4. Random sampling

5. Stratified sampling

Volunteer sampling

Volunteer sampling is when researchers post an advertise and wait for people to volunteer.

So, the pros of volunteer sampling are...

it is easy and can reach a lot of people

So, the cons of volunteer sampling are...

the sample is not very representative of the whole population

Opportunity sampling is when researchers find participants by...

approaching people who are willing and easily available.

The pro of opportunity sampling is that…

it’s a quick and easy way to sample participants.

However, the con of opportunity sampling is that the sample won’t be very

representative of the entire population

systematic sampling?

A list of everyone in the population which follows a system to pick every nth person.

However, one con of systematic sampling is…

It is difficult because the researcher has to obtain a list of the whole population.

A second con of systematic sampling is that

if there is a bias in how the data is listed, the sample may not be representative of the population.

Random sampling is…

Picking randomly from a list of the entire population, so that everyone has an equal chance of being a participant.

The pro of random sampling is that it is…

representative because every member of the population has an equal chance of being included.

However, a con of random sampling is that…

it’s more difficult because you need a list and a way to randomise it.

A second con of random sampling is that…

It still doesn’t guarantee that the sample will look like the population.

Stratified sampling

the sample has the same proportion of each subgroup as the total population.

3 Steps of stratified sampling :

1. Identify the subgroups in the whole population

2. Identify how many people from each subgroup are needed to have the same proportion as the original population

3. Randomly sample from each subgroup

So, a con of stratified sampling is that researchers may miss out an important subgroup by mistake! And this means…

The sample may not be completely representative.

Another con of stratified sampling is that is it …

difficult and time consuming

The first ethics guideline for psychological research is informed consent, which means that…

Participants have given written agreement to take part in the study and they fully understand what the study involves

Deception…

is when participants are told things that aren’t true and it is deemed unethical by the British Psychology Society, unless strictly necessary.

One reason deception might be necessary is the risk that participants might guess what the experiment is investigating, if they were told everything it involved. Knowing what the experiment is investigating might lead to...

Demand characteristics

What is British Psychological Society (BPS)?

It comes up with the ethical guidelines that all researchers should follow in their studies.

The third ethical guideline is protection from harm, which means that..

researchers must make sure the participants are in no more physical and emotional harm than they would be in normal life.

The fourth ethical guideline is the right to withdraw, which means…

Participants have the right to stop the study at any point and the researcher must tell them they have this right.

The fifth ethical guideline is confidentiality, which means that…

The fifth ethical guideline is confidentiality, which means that researchers must keep the identity of their participants anonymous.

An experiment is a study that investigates…

cause and effect relationships of two variables

What is a laboratory experiment?

Where the IV is manipulated in an artificial controlled environment

Pro of a laboratory experiment:

• High internal validity because the researcher can control extraneous variables which allows it to establish a valid cause and effect relationship

Con of a laboratory experiment:

• Due to an artificial setting the participants behaviour may not generalise to behaviour in everyday life. This means that a lab experiment can have low ecological and low external validity.

A field experiment is…

an experiment where the independent variable is manipulated in a natural, everyday life setting.

A pro of a field experiment:

they have higher ecological validity than laboratory experiments.

A con of a field experiment is …

it is harder to establish clear cause and effect relationships between the variables as they have less control over extraneous variables meaning the study has lower internal validity.

A quasi experiment is an experiment...

where the independent variable can’t be manipulated and randomly allocated to groups because it’s a personal characteristic of the participants.

Pro of a Quasi experiment:

Enable the researcher to investigate the effect of personal characteristics on behaviour

Con of a Quasi experiment:

The study may lack internal validity making it harder to establish cause and effect because participants can’t be randomly allocated to groups, meaning that uncontrolled participant variables might act as confounding variables.

A pro of the natural experiment is…

It enables researchers to study independent variables that it would be unethical or impractical to investigate in a laboratory or field experiment.

Con of a natural experiment …

No control over extraneous variables or confounding variables which reduces internal validity and makes it difficult to establish a cause and effect relationship.

Matched pair design:

the first type of experimental design is the Matched pairs design, which is when the researchers match the participants from the control and experimental groups for a particular participant variable .

Strength of a matched pair design:

By matching for participant variables we control these variables which increases the internal validity of the study

2 limitations of the matched pair design are:

• Time consuming for researcher

• Not possible for the researcher to match every participant on every possible participant variable

Independent groups design is…

when the researcher allocates different participants to each group but doesn’t match the participants for any variables.

Repeated measures design is….

when the same participants take part in all of the experimental conditions.

Limitations of a repeated measures design

• Participants have more of a chance to find out the true aim of the study - demand characteristics

• The results can be influenced by order effects which is when the order of experimental conditions influences the results of a study