Research Methods

• Research method: The process of collecting information about a theory or idea.

• Data collection method: The process by which information is directly collected.

• Experiment: A study in which an independent variable is manipulated to investigate its effects on a dependent variable. 

• Observation: A study in which there is no independent variable but data is simply collected by seeing what happens.

• Self-report: A study in which the participants are asked to state or explain their own feelings, opinions, behaviours or experiences, e.g. interviews or questionnaires.

• Directional hypothesis: States the direction of the predicted difference or relationship.
  • Typically used by psychologists when the findings of previous research suggest a certain outcome. 

• Non-directional hypothesis: Does not state the predicted direction of the outcome (more vague)
  • Typically used by psychologists when there is no previous research or if earlier research is contradictory.

• Null hypothesis: No difference is predicted.

• Hypothesis: A clear, precise, testable statement that states the predicted relationship between variables to be investigated. This is stated at the outset of any study.

• Independent variable: Changed or manipulated by the researcher.

• Dependent variable: Measured or observed by the researcher.

• Controlled variable: Stays constant throughout the experiment.

• Operationalisation: Clearly defining variables in terms of how they can be measured, e.g. intelligence is measured by test scores.

• Laboratory experiment: An experiment that takes place within a controlled environment within which the reserearcher manipulates the IV and records the effect on the DV, while maintaining strict control over extraneous variables.
  • Strengths: High interal validity due to control over extraneous variables; replication is more possible due to high levels of control which increases validity; usually standardised procedure.
  • Limitations: Hard to generalise leading to low external validity and mundane realism; demand characteristics as participants are usually aware that they are being watched.

• Field experiment: An experiment that takes place in a natural setting within which the researcher manipulates the IV and records the effect on the DV.
  • Strengths: High mundane realism and ecological validity; may have high external validity as participants may be unaware that they are being watched.
  • Limitations: Establishing cause and effect relationships is difficult due to lack of control over extraneous variables; precise replication is often impossible; ethical issues arise due to lack of consent and deception.

• Natural experiment: An experiment where the change in the IV is not brought about by the researcher, but would have happened even if the researcher had not been present. The researcher records the effect on the DV.
  • Strengths: Provides opportunities for research that would not otherwise be undertaken due to practical or ethical issues; high external validity. 
  • Limitations: Rarity of research event reduces scope for generalisation; lack of independent groups design means that the researcher may be less sure whether the IV affected the DV. 

• Quasi-experiment: A study that is almost an experiment but lacks key ingredients. The IV has not been determined by anyone- the variables simply exist, such as age. Strictly speaking, this is not an experiment.
  • Strengths: Are often carried out under controlled conditions so they share the same strengths as lab experiments (high internal validity; replication more possible).
  • Limitations: There may be confounding variables as quasi-experiments cannot randomly allocate participants to conditions.

• Experimental design: The different ways the testing of participants can be organised in relation to the experimental conditions.
• Independent groups design: Participants are allocated to different groups where each group represents an experimental condition.
  • Participant variables may impact outcome of impact on DV but can be tackled using random allocation; less economical than repeated measures design as participants contribute to a single result only; order effects are not a problem which reduces impact of demand characteristics.
• Repeated measures design: All participants take part in all conditions, ensuring that "like for like" is compared.
  • order effects are tackled with counterbalancing; order effects may invalidate findings as participants may improve or become bored; participants are more likely to figure out the aims of the study, leading to demand characteristics; participant variables are controlled and fewer are needed. 
• Matched pairs design: Participants are first matched on some variable/s that may affect the DV. Next, one member is assigned to Condition A and the other to Condition B.
  • Order effects and demand characteristics are less of a problem as participants only take part in one condition; there will always be some important difference between participants within a matched pair that may affect the DV; matching may be time-consuming and expensive.

• Extranous variables: Any variable, other than the IV, that may have an impact on the DV if not controlled.

• Confounding variables: Any variable, other than the IV, that may have affected the DV so we cannot be sure of the source of change to the DV.

• Demand characteristics: Any cue from the researcher, or from the research situation itself that may be interpreted as revealing the purpose of the investigation. This may cause participants to change their behaviour. 

• Order effects: In a repeated measures design, a confounding variable arising from the order in which conditions are presented, e.g. practice or boredom. 

• Investigator effects: Any effect of the investigator's behaviour on the research outcome, e.g. participant selection, tone or wording influencing answers etc. 

• Participant variables: Characteristics of individual participants that might influence the outcome of the study.

• Random allocation: An attempt to control for participant variables in an independent groups design which ensures that all participants have an equal chance of being in all conditions.

• Counter balancing: An attempt to control for the effects of order in a repeated measures design- half the participants experience conditions in one order, and the other half in another order.

• Randomisation: The use of chance in order to control for the effects of bias when designing materials and deciding the order of conditions.

• Standardisation: Using exactly the same formalised procedures for all participants in a research study.

• Validity: The extent to which an observed effect is genuine- does it measure what is was supposed to measure, and can it be generalised beyond the research setting?

• Internal validity: Whether the effects observed in an experiment are due to the manipulation of the IV and not other factors, e.g. demand characteristics.

• External validity: Relates more to factors outside the investigation, such as generalisation.

• Face validity: A measure of validity in which research is scrutinised to determine whether it measures what its supposed to.

• Concurrent validity: The extent to which a measure relates to another (meta-analysis)

• Ecological validity: The extent to which findings from a research study can be generalised to other settings and situations.

• Temporal validity: The extent to which findings from a research study can be generalised to other time periods and eras.

• Mundane realism: Is the experiment like real life?

• Reliability: Refers to how consistent the findings from a measuring device or investigation are.

• Test-retest reliability: A method of assessing the reliability of a questionnaire or psychological test by assessing the same person on two separate occassions (not too close together or far apart). This shows consistency and reliability. 

• Inter-observer reliability: The extent to which there is an agreement between two or more observers involved in observations of behaviour. This is done by correlating the results of two or more observers.

• Random sample: All members of the target population have an equal chance of being selected. People within the sample are given a number and then numbers are drawn using a lottery method.
  • Strengths: Free from researcher bias
  • Limitations: Difficult and time consuming; may still be unrepresentative; refusal to take part may leave you with a volunteer sample. 

• Systematic sample: Every nth number of the population is selected. A chosen interval is nominated or may be selected randomly to reduce bias. 
  • Strengths: Fairly representative; avoids researcher bias.
  • Limitations: Refusal to take part may leave a volunteer sample.

• Stratified sample: The composition of the sample reflects proportions of sub-groups in the target population.
  • Strengths: Avoids researcher bias; representative; generalisation more possible.
  • Limitations: Refusal to take part may leave a volunteer sample; may not be 100% representative as the strata cannot fully recognise how people are different.

• Opportunity sample: Anyone who happens to be willing and available to participate when asked.
  • Strengths: Saves time, money and effort.
  • Limitations: Unrepresentative so findings can't be generalised; volunteer bias may attract a specific type of person; subject to researcher bias due to control over who is asked; refusal to take part may leave you with a volunteer sample.

• Volunteer sample: Participants select themselves to participate.
  • Strengths: Easy and less time-consuming.
  • Limitations: Volunteer bias limits generalisation.

• Following BPS guidelines (respect, competence, responsibility and integrity)
  • Implemented by ethics committees in research institutions who often use a cost-benefit approach to determine whether research is ethical.

• Informed consent
  • Verbal consent or a signed consent form

• Deception and protection from harm
  • Debriefing; making participants aware of what the research will be used for and their right to withhold data.

• Privacy and confidentiality
  • Reminder of data protection; annonymity; initials or numbers used rather than names; holding no personal details.

• Presumptive consent: Consent is gained from a similar sample, so consent is presumed.

• Prior general consent: Participants essentially consent to being decieved as they agree to a number of studies. 

• Retrospective consent: Participants are asked for consent during the debriefing after the research has taken place.

• Pilot study: Small scale version of the experiment that takes place before the real thing is conducted. This allows the researcher to identify and modify any potential issues.

• Single-blind procedure: Participants are not made aware of the aims or condition of the study in order to control for confounding variables.

• Double-blind procedure: Neither the participants or the researcher is aware of the aims of the investigation.

• Control groups and conditions: If the change in behaviour of the experiment group is significantly greater than that of the control group, the researcher can conclude that the cause of this effect was the IV.

• Naturalistic observation: Watching and recording behaviour in the setting in which it would normally occur.
  • Strengths: High external validity; can be generalised to everyday life; high mundane realism
  • Limitations: Replication is difficult due to lack of control; uncontrolled extraneous variables make judging patterns of behaviour difficult.

• Controlled observation: Watching and recording behaviour within a structured environment, i.e. one where some variables are managed.
  • Strengths: Replication is easier due to extraneous variables playing less of a part.
  • Limitations: May be harder to generalise to a real-life setting

• Overt observation: Participant's behaviour is watched and recorded with their knowledge and consent.
  • Strengths: More ethical due to consent
  • Limitations: Participant reactivity reduces validity.

• Covert observation: Participant behaviour is watched and recorded without their knowledge or consent.
  • Strengths: Removes problem of participant reactivity which increases validity.
  • Limitations: Lack of consent makes this unethical

• Participant observation: The researcher becomes a member of the group whose behaviour they are watching and recording.
  • Strengths: Increased insight increases validity.
  • Limitations: "Going native" can reduce objectivity

• Non-participant observation: The researcher remains outside of the group whose behaviour they are watching and recording. 
  • Strengths: Remaining at an objective distance reduces the risks of "going native".
  • Limitations: Reduced insight can reduce validity.

• Unstructured observation: Writing down everything you see; very detailed; may be appropriate for small-scale observations.
  • Strengths: Richness and depth of data
  • Limitations: Qualitative data is harder to analyse; greater risk of observer bias due to lack of objective behavioural categories; researcher may not record all behaviour which reduces usefulness.

• Structured observation: Observations are quantified using a pre-determined list of behaviours and sampling methods; suitable for busy or large-scale observations.
  • Strengths: Quantitative data is easier to analyse; more straight forward and systematic; lower risk of observer bias.

• Behavioural categories: When a target behaviour (e.g. affection) is broken up into components that are observable and measurable (e.g. hugging or holding hands)

• Event sampling: A target behaviour or event is first established and then the researcher records it every time it occurs.
  • Useful for infrequent target behaviours or events; may overlook details for complex events.

• Time sampling: A target individual or group is established and then the researcher records their behaviour in a fixed interval of time, e.g. every 60 seconds.
  • Effective at reducing the number of observations that have to be made; sampled behaviour may be unrepresentative of whole observation.

• Strengths: Cost-effective; can be completed without a researcher which further reduces costs; can produce quantitative data which is easy to analyse; easy to cross-reference questions; can reach a large sample.

• Limitations: Social desirability bias (wanting to paint yourself in a positive light) reduces validity; aquiescence bias (agreeing with questions) reduces validity; response bias (responding in the same way for all questions) reduces validity; researcher bias; questions may be misunderstood; participants can lie which reduces validity.

• Unstructured interview: Works like a conversation with a loose topic criteria, allowing the interviewee to elaborate on their responses.
• Structured interviews: Pre-determined questions that are asked in a set order.
• Semi-structured interview: There is a pre-set list of questions but interviewers are free to ask follow-up questions, e.g. in a job interview.
• Focus group: A group discusses views, opinions and/or experiences, e.g. in market research or politics.

Evaluation:

• Strengths: High validity; interviewer can explain questions; structured interviews reduce differences between interviewers and are easy to replicate due to standardised format; can ask about more sensitive topics; detailed responses.
• Limitations: Inability to elaborate in structured interviews may cause frustration; analysis is time-consuming and difficult; interviewer and social desirability bias may invalidate findings;  hard to check for reliability; method of recording answers may damage trust and validity; requires a skilled interviewer.

Types of closed questions:

• Likert scale
• Rating scale
• Fixed choice question

Designing interviews:

• Interview schedule = List of questions to cover
• Schedule should be standardised to reduce interviewer bias
• Confirmation of confidentiality 
• Quiet and private setting

Writing good questions:

• Don't overuse jargon
• Try not to use emotive or biased language
• Avoid double-barrelled questions
• Avoid double negatives

• Correlation: An investigation into association between co-variables; recorded on a scattergram; lack of cause and effect relationship establishment means that this is not an experiment as correlation does not imply causation.
• Correlational hypothesis: A hypothesis of how the relationship between co-variables will occur.
• Positive correlation: As one co-variable increases, so does the other.
• Negative correlation: As one co-variable increases, the other decreases.
• Zero correlation: No correlation between co-variables.

Evaluation:

• Strengths: Useful for preliminary research as they provide a precise and quantifiable measure of how variables are related; relatively quick and economical; secondary data can be used which makes the process quicker.
• Limitations: Cannot demonstrate cause and effect relationships; doesn't account for intervening variables; can be misinterpreted

• Peer review: The assessment of scientific work by others who are specialists in the same field to ensure that any research intended for publication is of high quality. These experts should be objective and unknown to the researcher or author.

• Main aims of peer review: To allocate research funding; validate quality and relevence of research; to suggest ammendments or improvements. 

• Evaluation: Annonymity could be abused; publication bias may create a false impression of psychology as research is disregarded; ground-breaking research may be buried if you go against the status quo which slows down advancement.

• Qualitative data: Data that is expressed in words and is not numerical.
  • Strengths: Rich in detail; more insightful; greater external validity.
  • Limitations: Difficult to analyse; may be sujective and biased.

• Quantitative data: Data that can be counted or statistical.
  • Strengths: Simple to analyse and compare data; more objective; less open to bias. 
  • Limitations: Lower external validity and mundane realism due to less insight.

• Primary data: Information that has been obtained first-hand by the researcher for the purpose of the research project, e.g. observation or questionnaire.
  • Strengths: Targets necessary information; authentic
  • Limitations: Time consuming; requires effort.

• Secondary data: Information that has already been collected by someone else and so pre-dates the current research project, e.g. the work of another psychologist or government statistics.
  • Strengths: Inexpensive; low effort required.
  • Limitations: May be innacurate, low quality, irrelevant in some places or incomplete.

• Meta-analysis: "Research about research", refers to the process of combining results from a number of studies on a particular topic to provide an overall view. This may involve a qualitative review of conclusions and/or quantitative analysis of the results to produce an effect size.
  • Strengths: Allows data to be viewed with more confidence and results can be generalised across larger populations.
  • Limitations: Prone to publication bias as the researcher may leave out negative studies.

• Descriptive statistics: The use of graphs, tables and summary statistics to identify trends and analyse sets of data. 

• Measures of central tendency: The general term for any measure of the average value of a set of data, e.g. mean, median and mode.
  • Mean is the most sensitive and representative, but can easily be distorted by extreme values.
  • Median is not affected by extreme values, but is less representative.
  • Mode is easy to calculate but unrepresentative.

• Measures of dispersion: The general term for any measure of the spread or variation in a set of scores.
  • Range is easy to calculate, but can be unrepresentative.
  • Standard deviation is precise but can be distorted by extreme values.

• Standard deviation: A sophisticated measure of dispersion in a set of scores. It tells us how much scores deviate from the mean by calculating the difference between the mean and each score. All differences are added up and divided by the number of scores. This gives the varience. Standard deviation is the square root of the varience.
  • Low SD = Close to mean
  • High SD = Results were dispersed

1) Calculate the difference between the mean and each score.

2) Differences are added up and divided by the number of scores. 

3) This is the varience.

4) Standard deviation is the square root of the varience.

• Normal distribution: A symmetrical spread of frequency data that forms a bell-shaped pattern. Median and mode are all located at the highest peak.

• Skewed distribution: A spread of frequency data that is not symmetrical, where data clusters to one end.

• Negative skew: A type of distribution in which the long tail is on the negative (left) side of the peak and most of the distribution is concentrated on the right.

• Positive skew: A type of distribution in which the long tail is on the positive (right) side of the peak and most of the distribution is concentrated on the left.