Research Methods

  • Created by: msahay
  • Created on: 15-08-19 10:40

The Experimental Method

Research method using random allocation of participants and the manipulation of variables to determine cause and effect.

Variable - a factor that can be changed/measured e.g. intelligence

  • Independent variable (IV) = the factor manipulated by the researcher in an investigation
  • Dependent variable (DV) = the factor measured by the researcher in an investigation 

Researcher manipulates the independent variable (IV) in order to measure its effects on the dependent variable (DV). The effect of varying hours of studying (IV) on test scores (DV). 

If there is a significant effect on the DV due to the IV's effect, we can establish a cause-effect relationship. 

Operationalisation - expressing a variable into a measurable factor e.g. numerically. Important for reliable findings and replication in order to check validity. Intelligence can be operationalised through an IQ test. 

1 of 16

Extraneous & Confounding Variables

Extraneous variables - variables other than the IV that might affect the DV e.g. participant variable such as intelligence. 

Confounding variables - uncontrolled extraneous variables that negatively affect results by having an effect on the DV. 

Types of extraneous variables: 

  • Participant variables - e.g. age and intelligence of participants. 
  • Situational variables - the experimental setting and surrounding environment e.g. time of day, noise or temperature of room. 
  • Experimenter variables - changes in the personality, appearance and conduct of the researcher e.g. female researchers may obtain different results to male ones. 

If researchers investigated the effect of background music or silence on homework performance  using 2 classes, they would have to control a number of extraneous variables like age, homework difficulty. If participants in condition 1, were more intelligent than those in condition 2, researchers could not be sure whether differences in homework performance were due to the prescence of music or intelligence. Intelligence is the confounding extraneous variable. 

2 of 16

Demand Characteristics & Investigator Effects

Demand characteristics - participants guess the purpose of research, its aims or hypotheses and alter their behaviour because of this fact. 

  • they may try to please researcher by giving what they believe are"right" answers 
  • they may try to annoy researcher by giving what they believe are "wrong" answers - the screw you phenomenon. 
  • they may act unnaturally out of nervousness or due to fear of evaluation. 
  • they may act unnaturally due to social desirability bias - giving answers they believe are morally correct, rather than answering honestly. 

Investigator effects - a researcher's features influence participants' responses

  • Physical characteristics of the researcher - e.g. gender, age or ethnicity. Male participants may be less willing to admit sexist views to female researchers. 
  • Accent or voice of researchers - e.g. participants may respond differently to a researcher with a stern voice. 
  • Researchers may be unconsciously biased in their own interpretations of data and find what they expect to find. 
3 of 16

Laboratory Experiments

experiment conducted in a controlled environment, allowing the establishment of causality. 

  • standardised procedures and participants are randomly allocated 


  • High degree of control - experimenters control all variables, IV and DV are operationalised and measured for greater accuracy and objectivity --> leading to the establishment of causality
  • Replication - other researchers can repeat the experiment to check for validity and relability of results. 


  • Low ecological validity - high degree of control makes experimental situations artificial. This means that it can be difficult to generalise findings to other settings. 
  • Demand characteristics - participants are aware that they are being tested so they may unconciously alter their behaviour/response.
4 of 16

Field Experiments

experiment conducted in a naturalistic environment where the researchers manipulate the independent variable

i.e. they occur in real world settings rather than the lab, where naive participants are exposed to a set up social situation


  • High ecological validity - as the experiment is conducted in a real world setting, findings relate to everyday behaviour and can be generalised to other settings. 
  • No demand characteristics - as participants are naive, there are no demand characteristics exhibited by them 


  • Less control - as the study is not conducted in controlled conditions, it is more difficult to control extraneous variables so, causality is harder to establish. 
  • Ethical violations - participants are not aware that they are in an experiment, meaning that there is a lack of informed consent, especially in field experiments. 
5 of 16

Natural & Quasi Experiments

Natural experiment - experiment where the independent variable varies naturally 

Quasi experiment - where the researcher is unable to freely manipulate the independent variable or randomly allocate participants to different conditions.


  • High ecological validity - as the experiment is conducted in a real world settingfindings relate to everyday behaviour and can be generalised to other settings. 
  • No demand characteristics - as participants are naive, there are no demand characteristics exhibited by them 


  • Less control - as the study is not conducted in controlled conditions, it is more difficult to control extraneous variables so, causality is harder to establish. 
  • Replication - since the conditions will never be exactly the same again it is difficult to exactly repeat natural experiments to check findings for validity and reliability. 
6 of 16

Observational Techniques

Observations involve watching and recording behaviour, either in a naturalistic or controlled environment

  • Participant observation - the researcher is actively involved in the situation e.g. Zimbardo
  • Non-participant observation - the researcher is not actively involved in the situation

Overt observation - participants are aware they are being observed

  • Participants may show demand characteristics e.g. social desirability bias and therefore, act unnaturally, which reduces the validity of findings. 

Covert observation - participants are unaware they are being observed

  • High ecological validity - participants are unaware they are being observed so they act naturally and findings can be generalised to other settings. 
  • Practical method - can be used where deliberate manipulation of variables would be impractical
  • Ethical violations - if participants are unaware they are being observed, this can raise concerns about informed consent. 
  • Observer bias - researchers may see what they want to find in their results. 
7 of 16

Observational Design

Naturalistic observations - surveillance and recording of natually occuring events

Behavioural categories - dividing target behaviours into subsets of behaviours through the use of coding systems 

Codes are designed by the observers beforehand on a coding sheet to record the studied behaviour e.g. in a driving behaviour study, D = distracted - this would be an observed behaviour code. 

This allows for structured observations that are more objective and more easily quantifiable.

Sampling procedures

  • Event sampling - counting the number of times a behaviour is exhibited in an individual or target individuals. 
  • Time-sampling - counting the behaviour in a set time frame e.g. recording what behaviour is being exhibited every 30 secs. 

Inter-observer reliability - independent observers consistently code behaviour in same way. 

8 of 16

Self-Report Techniques

Self-report techniques - participants give information about themselves without researcher interference. 

Questionnaires - self-report method where participants record their own answers to a pre-set list of questions. 

  • Closed questions - involve yes/no answers or a range of fixed responses e.g. do you eat meat? 'always', 'usually', 'sometimes' or 'never'. Easy to quantify but restricts participants' answers. 
  • Open questions - allows participants to answer in their own words. They produce a greater depth and detail + freedom of expression for the participant but it is harder to analyse this qualitiative data. 

Interviewsself-report method where participants answer questions face-to-face with researcher. 

  • Structured interview - identical pre-set list of questions being asked to participants with researcher writing down their answers. 
  • Unstructured interview - an informal discussion of a topic between a researcher and participant, with follow-up questions being asked. 
9 of 16

Questionnaires Evaluation


  • Questionnaires are quick and cheap - large amounts of information from a large sample can be obtained relatively quickly and cheaply. 
  • No investigator effects - questionnaires can be completed without researcher interference, so investigator effects will not affect the responses given by participants. 
  • Replication - questionnaires use standardised questions so can be replicated to check for validity and reliability - especially if closed-questions were used. 


  • Questionnaires are subjective - the participant's perception of themselves may be quite different to how everyone else views them so their answers are subjective. 
  • Misunderstanding - participants may misinterpret questions e.g. what is meant by how often do you 'usually' do your homework?
  • Biased samples - participants that have the time to complete questionnaires may be a specific type of person and not actually representative of the whole population. 
  • Low response rates - questionnaires usually get a very low response rate 
10 of 16

Questionnaire Construction

When constructing a questionnaire....

Questions must be clear, concise, non-ambiguous and understandable by all participants - questionnaires should be tested using a pilot study on people who can provide feedback on the design. 

Use measurement scales - statements that assess participants agreement or disagreement and give a picture of their overall attitude

Aims - having an exact aim of the study should help to direct questions 

Using a previously successful questionnaire can be used as a basis for the questionnaire. 

11 of 16

Interviews Evaluation


  • Interviews can provide rich, qualitative data about the participants' subjective view of an issue with the use of open-ended questions. This can be complemented by any quantitative data produced by structured interviews. 
  • Structured interviews are standardised so they can be replicated to check the reliability of findings, though this may be more difficult with unstructured interviews. 
  • Complex issues - complicated issues tackled face-to-face with a researcher can help participants feel calmer and more relaxed. 


  • Demand characteristics - participants may display social desirability bias and give answers which they think are 'correct' (socially-appropriate) instead of answering honestly. 
  • Investigator effects - the gender, age or ethnicity may affect the participant's responses e.g. male partcipants may not express sexist views due to a female interviewer. 
  • Interview training - in order to carry out unstructured interviews, interviewers must be properly trained which can be difficult to achieve. 
12 of 16

Correlational Studies

measure the strength and direction of relationships between co-variables e.g relationship between locus of control and obedience.When talking about correlational studies always refer to co-variables, not IV and DV. 

  • Positive correlation - as one co-variable increases, the other one increases e.g. as stress increases, illness increases. 
  • Negative correlation - as one co-variable increases, the other decreases e.g. as age increases, dependency on mother decreases. 

Hypotheses for correlational studies:

Hypotheses for correlational studies predict a relationship between the 2 co-variables. They can be non-directional or directional (depending on whether past research indicates whether we should expect to find a relationship). 

  • 2-tailed non-directional experimental hypothesis - "there will be a correlation between stress and illness."
  • 1-tailed directional experimental hypothesis - "there will be a positive correlation between stress and illness."
13 of 16

Correlational Analysis

correlational study involves measuring the relationship between 2 co-variables e.g. stress and illness. 

co-variables must be operationalised and then plotted onto a scattergram which can indicate the direction and strength of the relationship between them. 

Statistical analysis of correlational studies data produces a correlational coefficient - a number between -1 and +1 which will indicate the exact direction and strength of the relationship between the 2 co-variables. 

E.g. correlational coeffecient 

  • +0.93 - strong positive correlation 
  • -0.46 - weak negative correlation
  • +0.04 - weak positive correlation
14 of 16

Correlational Analysis Evaluation


  • Allows predictions to be made - predictions can be made from correlatios e.g. predicting the number of ice-creams sold on hot days. 
  • Allows quantification of relationships - statistical analysis of correlational data can show the direction and strength of correlational relationship in quantitative terms
  • Correlational studies are a valuable preliminary research tool - they allow us to investigate relationships between two variables that we may decide to investigate in more detail through experimentation. 


  • Does not tell us about cause and effect - correlational studies are not done under controlled conditions, so they do not show causality. 
  • Quantification problem - correlations that appear low can sometimes be significant if the number of scores is high, while correlations that seem high are not always significant. 
  • Only works for linear relationships
15 of 16

Case Studies

 in depth, detailed investigations of one individual or a small group.


  • Rich detail - case studies provide depth and understanding about individuals and have a feeling of 'truth' about them. 
  • Only possible method to use - they allow psychologists to study unique behaviours or experiences that could not have been studied any other way e.g. privation effects. 
  • Can be used to challenge a theory about behaviour - just one unique case study can go against a proposed theory of behaviour.


  • Relies heavily on memory - case studies depend on participants having full and accurate memories, which is usually not the case. 
  • Researcher bias - researchers conducting case studies may be biased in their interpretations, making findings suspect. 
16 of 16


No comments have yet been made

Similar Psychology resources:

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