Ethics
Methods
Mind and Body
Nature vs Nurture
Planning Research
Representing Data

Other Sections
Cognitive Psychology
Gender
Mental Health
Social Psychology

Planning a Psychology Investigation

In this page will we look at the things you need to consider in planning a Psychological study, these techniques can be applied to any social science.

Aims and Hypotheses

It is important to know these are different and what these differences are.

The aim of a study is what you want to find out. When writing a project it is important that this is very specific.

A hypothesis states what your prediction is. There are two types of hypotheses. A null hypothesis where you state there will be no change. It is what you assume to be true until you can prove otherwise. The alternative hypothesis is the prediction you (usually) are trying to prove.

The below table shows an example of these to help you understand.

AimTo investigate how useful www.scienceaid.co.uk is for science students.
Null HypothesisScience Aid has no effect on students performance in tests.
Alternative HypothesisScience Aid improves the performance of students in tests.

Sampling

Sampling refers to the way that you select people for your investigation. There are several different methods that can be used. The table below outlines some of the different types and discusses the pros and cons of them.

Sampling TypeDefinitionAdvantagesDisadvantages
OpportunityYou use the first people who fit your criteria.Easiest sampling method, very inexpensive.Doesn't use a variety of people that represent the population. Can be biased because only people from your own social group are included.
SystematicTake every 2nd 10th etc. personHopefully more representative, simple to do, has no bias.Because it is only based on chance there is no way of ensuring it is representative.
RandomEveryone in the target population has an equal chance of being selected.Hopefully more represntative and has no bias.Can be difficult to implement and therefore expensive.
StratifiedThe proportions of people in the sample are equivalent to the population (eg. 50% female).Very representative and therefore the findings of an experiment using this method can be applied to everybody more easily.Complicated to make a sample that represents everybody so could take a long time and be expensive.