(2027 Exam) Aims, Hypotheses & Sampling

Test yourself on (2027 Exam) Aims, Hypotheses & Sampling

Test your knowledge with free interactive questions on Seneca — used by over 10 million students.

Aims and Hypotheses

Each research study specifies aims and hypotheses. An aim is what it is trying to achieve, while a hypothesis is a specific prediction of what it will find.

Aim

A researcher usually states the aim of their study.
This involves saying what they are trying to achieve, or what the point of their study is.
Usually the aim is linked to a real-world purpose, i.e. a reason why it is important to find something out.
- For example, a researcher may state that they aim to find out the effect of caffeine on sleep.

Hypotheses

A hypothesis is different from an aim.
It involves making a specific prediction of what will be found, expressed in terms of a change in variables.
Usually the hypothesis is based on theories and on past research findings, i.e. there is a theoretical rationale for the hypothesis.
- For example, a researcher may state a hypothesis that consuming 200mg of caffeine will increase the length of time it takes people to fall asleep compared to having no caffeine.

Experimental vs alternative

In an experiment, the researcher’s main hypothesis is known as an experimental hypothesis. It is also referred to as H1.
In a non-experimental study, it is typically called an alternative hypothesis.

Null hypothesis

Most studies also clearly state a null hypothesis (sometimes referred to as H0).
This is a statement of what will be found if the experimental/alternative hypothesis is not supported by the results.

Directional hypothesis

A directional or one-tailed hypothesis predicts the direction in which change is expected to occur.
It is used when previous research has suggested the direction of change.
It is precise and uses words such as: faster/slower, bigger/smaller, more/less etc.
- e.g.Alcohol increases reaction times.

Non-directional hypothesis

A non-directional or two-tailed hypothesis simply predicts change and does not specify direction.
It is used when there is no previous research. It is non-specific and uses words like: effect, change, difference etc.
- e.g. Alcohol will affect reaction times.
NB. All variables must be fully operationalised e.g. alcohol measured in units; reaction times measured in seconds.

Populations and Samples

Any research study needs a group of participants. These are called the sample, and they are drawn from a wider group called the target population.

Sampling

Sampling means selecting a group of participants who will take part in the study.

Populations

A sample always comes from a broader population.
This does not necessarily mean the whole population of a country, but could be a specific group.
This is known as the target population.
- For example, all sixth-form school pupils in the country is an example of a target population, and a selection of 50 sixth-form school pupils is an example of a sample.

Representation

A key aspect of sampling is that the sample should be representative of the target population.
This means that they should have similar characteristics.
Studying a representative sample allows the researcher to generalise the findings to the target population. This is a key aim of any research.

Sampling Methods

There are multiple ways of obtaining a sample for a research study. Four major sampling methods are opportunity sampling, systematic sampling, volunteer sampling and stratified sampling.

Opportunity sampling

Opportunity sampling is the most common sampling method. It involves accessing participants on the basis of their convenient availability to the researcher:
- Examples of opportunity sampling include conducting research on the researcher’s own friends, classmates or students.
- Opportunity sampling is very prone to bias because the most easily available participants may not be representative of the target population.

Systematic sampling

Systematic sampling involves applying a regular system or rule when selecting participants:
- Examples of systematic sampling include picking every 50th person that walks along a corridor, or every 100th name in the phone book, or posting a questionnaire to every 10th house in a village.
- Systematic sampling reduces researcher bias, but some potential participants may be excluded e.g. because they are not in the phone book or do not live in a house. This leads to bias.

Volunteer sampling

Volunteer sampling allows participants to select themselves, such as by responding to an advert or email call for participants:
- Eg. posting an advert on a school noticeboard, asking people to complete your online survey.
- One source of bias with volunteer sampling is that certain personalities are more likely than others to come forward and help the research. This may have affected classic research studies such as Milgram’s obedience research.

Stratified sampling

Stratified sampling involves selecting participants in such a way as to recreate the same proportions of groups that exist in the population:
- An example of stratified sampling would involve selecting people from different ethnic groups to create a sample with the same proportions as exist in the target population.
- This reduces bias by making the sample more representative, but before stratification can occur, participants must already have been selected using another sampling method.

Random sampling

In random sampling like the National Lottery, all members of the target population must stand an equal chance of being selected.
Random sampling does not guarantee a representative sample, but the laws of probability predict that the chances of selecting a biased sample are minimal.
- E.g. putting the names of every member of the target population into a hat and pulling a sample out (without looking!).

Evaluation of random sampling

Strengths:
- If the sample is large enough, the rules of probability suggest that it should be representative.
Limitations:
- Participants may not be willing or able to take part in the research.
- Sample could still be biased in terms of variables such as gender, age, ethnicity etc.