Topic Summaries

Scientific processes

A-Level > Psychology > AQA > A-Level Psychology Topic Summaries > Research methods > Scientific processes
Previous Module
Next Module

Addiction

  • Aim: what the researchers are investigating and why. For example, in one of Asch’s variations, his aim was to study the effects of social support on rate of compliance.
  • Hypothesis: they aim to test a hypothesis.The hypothesis is what they predict will be the results of their study. The three different kinds of hypothesis are:
    • Directional: predicts the direction which the research will go in
    • Non-directional: predicts that there will be a difference between the variables but doesn’t specify a direction.
    • Null hypothesis: prediction that there is no relationship between two variables
  • Sampling:
    • Population: an entire group of people that the researcher wants to study for example all men, all teenagers, all UK adults.
    • Sample: a part a population which is representative of the entire population
    • Sampling techniques:
      • Random sampling: where the researcher selects a random sample of participants from a target population. This may be done by choosing names from a hat or from a random name generator. This method allows everyone to have a fair chance of being selected.
      • Systematic sampling: selects a sample of participants based on set intervals. For example, picking every 5th and 9th name from a list.
      • Stratified sampling: the researcher with split the population into different groups, working out what percentage of the population is in each group and then using random sampling to select names.
      • Opportunity sampling: the researcher will approach people that are available and ask them to take part in the study.
      • Volunteer sampling: this is where a researcher will put out an advertisement for volunteers, and people will sign up based on their own interest.
  • Pilot studies: a form of practice research using a downscaled group and running through the procedure. The aim of a pilot study is to identify any areas for improvement in the study design before fully conducting the research. For example, if the task is too difficult, the participants may lose interest in the study and won’t perform as the researcher desires. By detecting these issues, the researcher may be able to save time and money.
  • Experimental designs:
    • Repeated measures: where participants will participate both research conditions. They’ll complete task A, followed by task B and then the results are compared.
    • Independent groups: participants are divided into two groups and participate in separate conditions and then results are compared.
    • Matched pairs/independent group design: a group of participants are matched with another group of participants based on similar characteristics of the original group to try and control for those characteristics.
  • Observational design:
    • Behavioural categories: the researcher curates a list of behaviours to observe and record. This allows the study to be replicable and for different observers to be consistent in what they are looking for. Behaviour will then be coded into these categories rather than being written descriptively.
    • Event sampling: this is where the researcher will record every time they observe a behaviour
    • Time sampling: this is where the researcher will record behaviour at select intervals of time
  • Variables:
    • Independent variable: is something that is changed by the researcher.
    • Dependent variable: is the thing that is measured by the researcher.
    • Extraneous variables: unwanted variables that can lead to results of the experiment being skewed, which therefore leads to false conclusions being drawn.
    • Confounding variable: one that hasn’t been properly controlled and has actually affected the results of the experiment (i.e.something other than the IV that has changed the DV).
    • Operationalisation of variables: where the researcher defines a variable in terms of the study which allows them to turn an abstract concept into something measurable. For example, if the dependent variable the researcher is studying is anxiety, they will need to find a way to measure the level of anxiety, such as by using a blood pressure monitor.
  • Means of controlling experiments:
    • Random allocation: this is where people are randomly assigned to a group which helps to remove any researcher bias
    • Counter-balancing: used to deal with order effects in a repeated measure design. Order effects refer to how completing tasks in a specific order may affect the research (e.g.if someone completes one task, they may already think they have figured out the aim, which may affect performance in the second condition). Half of the participants will complete the study in one order, and the other half will complete it in the other order so as to combat order effects.
    • Randomisation: where participants are randomly allocated to conditions in order to eliminate bias.
    • Standardisation: where the conditions and standards are kept the same for every participant.
    • Control groups: a group of people who don’t take part in the experimental conditions and instead act as a group to compare the experimental group to.
  • Demand characteristics and investigator effects:
    • Demand characteristics: extraneous variable where a participant acts according to how they think the researcher wants them to, reducing ecological validity.
    • Investigator effects: an extraneous variable where the characteristic of the researcher affects the participants’ behaviour, leading to a reduction in ecological validity.
  • Ethics: when research is ethically conducted, health and safety of the participants will be considered. The British Psychological Association code of ethics covers:
    • Consent: participants should be told the aims of the study, data being collected, and any possible risks of the study before giving their informed consent. They have the right to withdraw at any time. The researchers must consider the participants actual ability to consent e.g. mental stability or undue pressure.
    • Deception: researchers can’t mislead participants about the nature of the study. But there are a few exceptions which require a level of deception (e.g. Asch’s study wouldn’t have provided valid results if participants knew the true nature of the study). It can be scientifically justified to have some degree of deception, so psychologists must use risk management to avoid harm:
      • General consent: telling the participants that they will be deceived without telling them how. This may affect the participants’ behaviour.
      • Retrospective consent: telling the participants that they were deceived after the research and then asking for consent to use their data.
    • Confidentiality: personal data of participants cannot be disclosed. Any identifiable data cannot be published.
    • Debriefing: after the research has been conducted, researchers have to explain all details of the study to the participants. If a participant has been harmed in the study, the researched must address the harm so that the participant is in a stable state to leave.
  • Peer review: assesses the scientific credibility of a research paper before it’s published in a research journal. Researchers will submit their paper to the journal who sends it to an expert to evaluate its scientific validity, then accept, revise, or reject the paper.
    • Single-blind: researchers don’t know the name of the reviewers.
    • Double-blind: the researcher doesn’t know the name of the reviewers and the reviewers don’t know the name of the researcher.
    • Open review: the researchers and reviewers are both known to one another.
  • Economic implications of psychological research:
    • Attachment research can inform policies like maternity/paternity leave to invest in the secure attachment of children and help them lead healthier, productive lives.
    • Memory research can inform criminology and the legal system, reducing wrongful conditions and saving money on appeals and prison costs.
    • Mental health research can inform our understanding of people’s holistic health, particularly in preventative and best practice measures that reduce strain on the NHS and increases people’s productivity in the long term.
  • Reliability:
    • Test-retest: presenting the same participants with the same test on two separate occasions, then see if there is a positive correlation between the two.
    • Inter-rater: to what degree different psychologists give similar estimates.
    • Inter-observer: do independent observations of the same situation at the same time tally every time? Comparing results from different observers can establish a correlation – if the correlation is 0.8 or more, it’s statistically significant.
  • Validity:
    • Temporal: are findings permanent/relevant to date?
    • Population: can it be generalised to the whole population?
    • Concurrent: do the results match an already established test?
    • Face: does the test do what it says it’s going to do?
    • Internal: does the study measure what it wants to?
    • External: can the results be generalised to other situations outside research environment?
    • Construct: does it successfully measure a concept it’s supposed to?
    • Ecological: is it generalisable to the real world?
  • Features of science:
    • Objectivity: observations made without judgement from a neutral perspective.
    • Empirical method: where research is carried out in a scientific manner based on empirical judgements.
    • Reproducibility: where experiments are replicated to ensure valid findings.
    • Falsifiability: when a hypothesis is tested there must be a possible observation that it can be proven false.
    • Paradigm shifts: science isn’t fully objective and unbiased, nor does it always progress linearly. Most scientists accept exiting theories as true and then use these theories to find supporting research, but may need to adjust if theories are challenged. A paradigm shift is where we replace one pre-existing set of ideas with a new one.

Unlock Scientific processes

Subscribe to SnapRevise+ to get immediate access to the rest of this resource.

Premium accounts get immediate access to this resource.

Previous Module
Next Module