Research Methods

HideShow resource information


  • Research questions arise from observations and psychologists use research to try to answer them
  • Research starts with an aim which is a broad statement about what is being investigated
  • Research uses H1 (research/experimental/alternative hypothesis) which predicts a difference or correlation and will be retained if evidence supports it. If H1 is rejected then the H0 (null hypothesis) is retained instead as this predicts no difference or no correlation
  • Experimental hypotheses contain an independant variable which will be manipulated by the experimenter. The independant variable is predicted to have an effect on a dependant variable which will be measured by the experimenter
1 of 14


  • The experimental method involves manipulating an IV to measure a change in the DV when all other variables are controlled
  • Extraneous variables are other variables (beside the IV) that could affect the DV if not controlled. Uncontrolled variables that could have affected the DV during the experiment are known as confounding variables
  • Unwanted variables can be controlled through standardisation, counterbalancing or randomisation
  • There are 3 types of experiment: laboratory, field and quasi
  • There are 3 types of experimental designs: repeated measures, independant groups and matched pairs design
2 of 14


  • Repeated measures designs have the problem of order effects when the others do not. Independant groups have the problem of participant variables which is reduced in matched pairs & eliminated by repeated measures. Matched pairs are more time-consuming than the other 2
  • The main strength of the experimental method is its ability to reliably establish cause & effect through high levels of control. However, controlling situations & other variables may distort reality & can result in a lack of ecological validity in the case of lab experiments
3 of 14


  • Variables are not directly manipulated in non-experimental methods. Instead, they investigate phenomena as they occur
  • Non-experimental methods may be used to increase validity and to avoid ethical problems associated with manipulating behaviour. They are also used when experiments are not practically possible
  • The main non-experimental methods are self report methods, observational studies, correlation studies, case studies and content analyses
4 of 14


  • Methods of self report require participants to answer questions
  • Questionnaires use pre-set questions which are normally written down and answered privately
  • Structured interviews also use pre-set questions, whereas unstructured interviews base questions on interviewees answers
  • Open questions allow participants to answer questions freely whereas closed questions offer fixed responses
  • Pilot studies are practice studies which allow researchers to trial questions and other procedures
  • The main strength of self-report methods is they give access to thoughts and feelings, but this relies on participants being able or prepared to express them
5 of 14


  • Observational methods simply involve watching and recording behaviour
  • Lab observations are carried out in controlled environments whereas natural observations are carried out in the field.
  • Covert observations are undercover whereas overt observations are open and participants know they are being observed
  • Participant observations involve the observer becoming involved in the group or situation they are observing, whereas in non-participant observations the observer just stands back and watches
  • The main strength of observations is that researchers can observe behaviour first-hand rather than having to trust self-reports. However, there are potentially problems with observer bias and the observer effects
6 of 14


  • A correlation measures the relationship between 2 variables
  • A positive correlation means 2 variables change in the same direction, a negative correlation means they change in opposite directions and a zero correlation means they show no pattern of relationship
  • Correlations do not allow researchers to establish cause and effect. However, they do allow them to statistically analyse naturally occurring events that could not be set up experimentally
7 of 14


  • Case studies are in-depth investigations of one-person or group of people
  • Case studies are detailed enough to give valid findings but it is difficult to generalise from them
  • Allow researchers to study events that they could not practically or ethically manipulate
  • Are efficient as it only takes 1 case study to disprove a theory
  • The researcher can become too involved and lose their objectivity. May misinterpret or influence outcomes
  • Often picked up after the event and so its hard to establish cause and effect
8 of 14


  • A content analysis involves interpreting secondary material
  • Quantitative analysis produces numerical data
  • Qualitative content analyses produce descriptive data
  • Content analyses allow researchers to study people they cannot access, but because of that, findings are more open to interpretation
9 of 14


  • Qualitative research produces descriptive data whereas quantitative research produces numerical data
  • Self-report methods, observations, case studies and content analyses can be used as part of qualitative research
  • Experiments, correlations, self-reports, observations and content analyses can be used as part of quantitative research
  • Qualitative research provides rich, valid data whereas quantitative research provides objective data that is easier to analyse
10 of 14


  • A population describes the entire group of people that research findings should be generalised to. Meanwhile, a sample is a group drawn from the population that a researcher actually studies and generalises from
  • Random sampling is when every person in the sampling frame has an equal chance of being selected
  • Systematic sampling iis when every nth person is selected from the sampling frame
  • Stratified sampling is when the sampling fame is split into different sub-groups and then people are selected from each sub-group in proportion to their occurence in the population
  • Opportunity sampling is when people are selected on the basis of convenience
  • Random, systematic and stratified sampling tend to give representative samples, although the 1st 2 techniques can occasionally result in 'freak' samples.
  • Opportuntiy sampling tends not to produce very representative samples but it is less time-consuming than the other methods
11 of 14


  • Ethical research is research that protects the welfare of participants
  • The main ethical guidelines covering psychological research are: consent, deception, debriefing, withdrawal from an investigation, confidentiality, protection of participants and working with animals
  • Unethical research can be justified if the costs are outweighed by the benefits of the findings
12 of 14


  • Graphs are used to represent quantitative data pictorially
  • Bar graphs are used to show differences in discrete data
  • Line graphs and histograms can both be used to show differences in continuous data
  • Scattergrams are used to show relationships between 2 sets of continuous data
13 of 14


  • A measure of central tendency describes the trend of data by calculating a typical or average value for a data set
  • The mode does this by identifying the most common score
  • The median finds the middle score
  • The mean calculates the arithmetic average by adding together all scores and dividing by the total number of scores
  • A measure of dispersion describes the spread of a set of scores (around the average)
  • A range does this by working out the different between the highest and lowest score
  • The standard deviation basically calculates the average by which each score deviates from the mean
14 of 14


No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Research methods and techniques resources »