Research Design

Aims

The main point of the study, normally TO INVESTIGATE e.g. if different types of music affects your maths grade

Hypothesis

testable prediction e.g calming music will increase your maths grade compared to rock music which will decrease your maths grade - vice versa

What are the types of experimental hypothesis?

Directional hypothesis → states the direction of the difference e.g. one condition will affect… more than the other

Non-directional hypothesis → Doesn’t specify a direction, just a difference. There will be a difference in maths grade depending on the type of music.

How to know which type of hypothesis to use?

• Directional - if there IS past research already to suggest the direction

• Non directional - if there is NO past research

Whats the difference between a population and a sample?

• population is everyone you are interested in

• sample: subset of a population that is used for the experiment

Whats the difference between a random and non-random sampling method?

• Random - every person has the same/equal chance of being picked and less likely to get a biased sample

• Non-random - every person does not have an equal chance + normally more convenient and cost effective but can lead to bias

What are sampling methods?

• Sampling methods are participants selected for a study form a sample.

• This is taken from a target population = all members of the group in which you are interested in (e.g. 5 year olds in the UK)

• We need to gain a representative sample (typical of the population in question) so that the findings will be generalisable, but this isnt always possible, it depends upon your sampling procedure.

• Inevitably the vast majority of samples contain some sort of bias.

What are the types of sampling methods?

• Random sampling

• Systematic sampling

• Stratified sampling

• Opportunity Sampling

• Volunteer sampling

Random sampling

Sophisticated form of sampling in which all participants of the target population have an equal chance of being selected.

Example: put every participants name in a hat and picked by random, picking from a hat

Evalutation of Random Sampling

+All members of the target population have an equal chance of being selected

+For very large samples it provides the best chance of unbiased representative sample - it is free from researcher bias, as researcher has no influence over who is selected.

- The sample can still be biased and unrepresentative e.g. if too small or if selected pps refused to participate.

Systematic sampling

Every nth member of the target population is selected and a sampling frame is produced using this system.

Examples: Every 3rd house on a street, every 5th pupil on a school register, pcik every nth person

Evalutation of Systematic sampling

+Unbiased as participants are selected objectively, so avoids researcher having any influence over who is chosen.

+Fairly representative - it would be difficult (though not impossible) to get an all male sample for example.

-Not truly biased/random unless you select a number using a random method and start with this person and then select every nth person.

Stratified Sampling

Researcher identifies the different types of people that make up the target population and work out the proportion needed for the sample to be representative (Strata).

Looking at the proportions and randomly selecting after you have calculated

• Example: In London, 80% of people support Millwall, 15% support Crystal Palace and 5% Charlton Athletic.

In a stratified sample of 20 pps, there would be 16 Millwall fans, 3 CP and 1 Charlton who would be randomly selected from all the fans of each team.

Evaluation of Stratified Sampling

+Likely to be more representative than other methods because there is both proportional and randomly selected representation of sub-groups. This allows for generalisation of the findings.

+Avoids researcher bias during selection due to random selection

-Time consuming to identify sub-groups, randomly select them and then contact them

-The identified strata cannot reflect all the ways that people are different, so complete representation is impossible.

Volunteer Sampling

Individuals who have chosen to be involved in a study. Also called self-selecting. Researcher must place an advert and wait for respondents.

Example: Putting posters up for others to see and if they are interested to volunteer, they’ll do it / go to them.

Evaluation for Volunteer sampling

+Relatively convenient and ethical if it - leads to informed consent

+You are likely to get full cooperation from pps as they have volunteered to participate, so are very willing

-May be biased for away, making people reconsider whether they want to volunteer.

Opportunity Sampling

Simply selecting those people that are available at the time. AKA convenience sample. It involves selecting the first people you meet who fit the right criteria,who you are prepared to ask for help and who are prepared to help you.

Example: People on highstreets may hand out leaflets to you.

Evaluation of opportunity sampling

+A good sized sample can be generated quickly, conveniently and in an economical way.

+It is perfectly adequate when investigating processes which are assumed to work in a similar way in all humans

-It can also be biased by the researcher, who is likely to only approach people who look ‘helpful’ / from their own social group.

-Unrepresentative as people who are not there dont have the opportunity to participate. E.g only people on the high street at the time have chance of being selected - making sample ungeneralisable