Research Methods

correlation - mathematical technique in which a researcher investigates an association between two co-variables, usually represented on a scattergram

correlation coefficient - a no. between -1 (negative correlation) and +1 (positive correlation) that represents the direction and strength of a relationship between two co-variables

strengths: provide precise and quantifiable measure of how two variables are related, and are quick and economical to carry out

weaknesses: does not establish causation - unknown which co-variable is causing the other to change, it may be a casual relationship but have nothing to do with one another (a third variable)

AO1:

- an in-depth investigation, description and analysis of a single individual, group or event

- qualitative data, usually in the form of questionnaires, interviews, observations or longitudinal studies

AO3:

strengths: rich, detailed insights, less likelihood of demand characteristics 

weaknesses: unable to generalise to wider society as case studies focus solely on the individual with a particular ilness/life situation etc. case studies may be inaccurate, participants may respond to questionnaires and interviews with bias to appear socially desirable, or they may suffer memory decay which alters the findings

AO1: a research technique that enables the indirect study of behaviour by examining communications that people produce e.g. in texts, emails, TV, films etc.

- coding (quantitative data) = stage of analysis in which the communication to be studied is analysed by identifying each instance of the chosen categories (words, phrases, behaviours etc.)

- thematic analysis (qualitative data) = involves identifying implicit of explicit ideas that are recurrent within the data. themes will often emerge once the data has been coded

AO3:

strengths - circumnavigates ethical issues as much of the content already exists within the public domain (thus there are no issues with obtaining permission etc.) produces both quantitative and qualitative data = flexible

weaknesses - danger that the reseacher may attribute opinions/attitudes to the speaker/writer that were not originally intended as it is studied outside of the original context. qualitative data lacks objectivity

reliability - refers to how consistent the findings from an investigation are. the more consistent the findings, the more reliable they are

test-retest reliability - assessing the reliability of a psychological study by assessing the same person on two seperate occasions (must be sufficient time between retests so participants do not recall their previous answers but not so long that their opinions and attitudes change). this shows to what extent the test produces the same answers (i.e. is it reliable)

inter-observer reliability - the extent to which there is agreement between two or more observers involved in observations of a behaviour. this is measured by correlating the observations of two or more observers. if total no. of agreement / total no. of oberservations is +.80, it has high inter-observer reliability

ways of improving - questionnaires: replace open ambiguous q's with closed ones / interviews: avoid leading q's by making it structured with closed q's / experiments: control any extraneous variables to make sure research is replicable / observations: behavioural categories must be operationalised and should not overlap. could also be a covert observation

validity - the extent to which an observed effect is genuine

face validity - basic form of validity in which a measure is scrutinised to determine whether it appears to measure what it is supposed to measure (assessment of validity)

concurrent validity - the extent to which a psychological measure relates to a existing or similar measure (assessment of validity)

internal validity - whether the effects observed in an experiment are due to the manipulation of the independent variable and not another factor

ecological validity - the extent to which findings from a research study can be generalised to other settings and situations (external)

temporal - the extent to which findings from a research study can be generalised to other historical times and eras

ways of improving - experimental: control groups, standardisation (e.g. single-blind/double-blind tests to reduce demand characteristics) / quesionnaires: lie scale to combat effects of social desirability bias / observations: covert and behavioural categories 

probability - a measure of the likelihood that a particular event will occur where 0 indicates statistical impossibility and 1 statistical certainty

significance - whether there is a difference or correlation. a significant result means the researcher can reject the null hypothesis 

critical value - the numerical boundarybetween acceptance or rejection of the null hypothesis

- one-tailed test if hypothesis is directional

- two-tailed test if hypothesis is non-directional

- p = 0.05

type I error - the incorrect rejection of a true null hypothesis (false positive)

type II error - the failure to reject a false null hypothesis (false negative)

Chi-squared - test of difference, unrelated design, nominal data

Sign test - test of difference, related design,  nominal data

Mann-Whitney - test of difference, unrelated design, ordinal data

Wilcoxon - test of difference, related design, ordinal data

Spearman's rho - test of association/correlation, ordinal data

Pearson's r - test of association/correlation, interval data

Related t-test - test of difference, related design, interval data

Unrelated t-test - test of difference, unrelated design, interval data

Unrelated design = independent groups / Related design = matched pairs and repeated measures / Nominal = categories (discrete, does not overlap) / Ordinal = continuous data, rating (subjective) / Interval = scales of measurments with accepted units of precisely defined sizes, unmanipulative data (e.g. centimetres, time etc.)

Abstract - key details of research report

Introduction - aims, hypothesis and a look at past research

Method - description of what the researcher did, including the design, sample, apparatus/materials, procedure, ethics

Results - description of what the researcher found, including descriptive and inferential statistics

Discussion - consideration of what the results of the study may tell us in terms of psychological theory

References - list of sources that are referred to or quoted in the article, e.g. journal articles, books, websites and their full details

paradigm - a set of shared assumptions and agreed methods within a scientific discipline 

paradigm shift - result of a scientific revoluion: a significant change in the dominant unifying theory within a scientific discipline

objectivity - where all sources of personal bias are minimised so as not to distort or influence research processes. lab experiments tend to have the most control and objectivity 

empirical method - scientific approached that are based on the gathering of evidence through direct observation

replicability - the extent to which scientific procedures and findings can be repeated by other researchers. determines the validity of the study and whether it can be generalised 

falsifiability - the principle that a theory cannot be considered scientific unless it admits the possibility of being proved untrue. the hypothesis should be able to be tested

- they work out the probability of whether a particular set of data could have simply occured by chance

- we may have found a difference between our two samples but we want to know if this difference is big enough to be of significance

independent groups - participants are allocated to different groups where each group represents one experimental condition. random allocation is used to combat this whereby participants are randomly allocated to the conditions

repeated measures - all participants take part in all conditons of the experiment. o combat order effects, counterbalancing is used (half the participants take part in condition A then B, then the other half do B then A

matched pairs - pairs of participants are first matched on variables that may affect the dependent variable (e.g. IQ) then one member of the pair is assigned comdition A and the other B

lab experiment - takes place within a highly controlled environment within which the researcher manipulates the IV and records the effect on the DV whilst maintaining strict control of extraneous variables. / strengths: high control thus replicable / weaknesses: low external validity so lacks generalisability. p's may respond to demand characteristics and has low mundane realism

field experiment - takes place in a natural setting within which the researcher manipulates the IV and records the effect on the DV. / strengths: high mundane realism, behaviour is more natural, less aware that they're being studies so high external validity / weaknesses: less control, ethical complications if participants are unaware they're being studied

natural experiment - where the change in IV is not brought about by the researcher but would have happened even if the researcher had not been there (natural setting) / strengths: high external validity / weaknesses: naturally occuring event may be rare so may limit the scope for generalising. p's may not be randomly allocated so researcher unsure whether IV has affected the DV

quasi-experiments - IV is based on existing difference between people (e.g. age and gender) and cannot be manipulated. / strengths: control / weaknesses: cannot randomly allocate p's

population - group of people who are the focus of the researcher's interest

sample - a group of people drawn from target population

sampling techniques:

- random sample (pick number out of hat) / strengths: free from researcher bias / weaknesses: time-consuming, unrepresentative

- systematic (every nth number) / strengths: avoids researcher bias, represenatative

- stratified (size of whole sample/size of population x size of layer) / strengths: avoids researcher bias, randomly selected / weaknesses: not representative of entire population

- opportunity (asking) / strengths: convenient, less time-consuming and cheaper / weaknesses: researcher bias, unrepresentative

- volunteer (p's volunteer) / strengths: convenient, less time-consuming and cheaper / weaknesses: attracts certain profile of people

ethical issues:

- informed consent (making p's aware of aim, rights and procedures, including right to withdraw)

- deception (deliberately misleading or witholding information)

- protection from harm (should not be put under stressed and should be debriefed at the end)

- privacy and confidentiality (p's have the right to control information about themselves and maintain anonymity)

BPS code of ethics - a quasi-legal document produced by the BPS that instructs psycholigists in the UK about what behaviour is and is not acceptable when dealing with p's. It is built around four major principles: respect, competence, responsibility, integrity.

pilot studies - small-scale version of an investigation that takes place before the real investigation is conducted. the aim is to check that the procedures, materials, measuring scales, etc. work and to allow the researcher to make any necessary changes

single-blind procedure - participant unaware of aim, research aware (attempt to control demand characteristics)

double-blind procedure - neither participant or researcher aware of aim (attempt to control demand characteristics and researcher bias)

naturalistic observation - watching and recording behaviour in the setting within which it would normally occur / strengths: high external validity as behaviour is natural / weaknesses: no control so lack of replicability and lack of control over extraneous variables

controlled - observing behaviour within a structured environment (controlling variables) / strengths: happens in a lab so variables are controlled, more replicable / weaknesses: lack of external validity, may respond with demand characterstics

covert - participants' behaviour is observed without their knowledge or consent (undercover) / strengths: avoids participant reactivity so increases validity of data / weaknesses: unethical due to lack of informed consent

overt - participants' behaviour is observed with their knowledge or consent / strengths: ethically acceptable / weaknesses: participant reactivity

participant observation - researcher becomes a member of the group whose behaviour they are observing / strengths: increased insight, increased validity / weaknesses: loses objectivity

non-participant - researcher remains outside of the group they are observing / strengths: maintains objectiveness / weaknesses: lose insight

structured observations (recording of data is easier and systematic and provides quantitative data which is better for comparing)

- behavioural categories (should not require further interpretation, all possible forms of the target behaviour must be included in the checklist and they should be discrete and non-over-lapping)

- event sampling i.e. counting the no. of times a behaviour occurs (useful when behaviour happens infrequently and could be missed if time sampling is used. however, if behaviour is too complex it may be overlooked

- time sampling i.e. recording behaviour within a pre-established time frame (reduces no. of observations that have to be made, but is unrepresentative of the observation as a whole)

unstructred observations (qualitative data which may be difficult to analyse, greater rich of detail but at more risk of researcher bias as may only record details that 'catches their eye'

- continuous sampling i.e. all instances of a target behaviour is recorded 

questionnaires - a set of written questions used to assess a person's thoughts and experiences. these may include open (qualitative) questions or closed (quantitative) questions / strengths: cost effective, can gather large amounts of data in short space of time, can be distributed to a large no. of people and researcher does not have to be present so little effort is recquired. if quantitative, data is easy to analyse / weaknesses: respondents may not be truthful, may try to appear socially desirable, may show response bias in which respondent always checks on particular box

interviews - a 'live' encounter where the interviewer asks a set of questions to assess a participant's thoughts and experiences. they may be structured (pre-determined set of questions) or unstructured (general conversation) or semi-structured (pre-determined questions with follow-ups). 

structured: easy to replicate, reduces difference between interviewers (so avoids response bias) however, it makes it harder to deviate or elaborate on topics

unstructured: flexible, gives insight, however it is hard to analyse and compare and p's may answer untruthfully to appear socially desirable

designing questionnaires:

likert scales - the respondent indicates their degree of agreement with a statement using a scale, with strongly agree at 1 and strongly disagree at 5

rating scales - respondents identify a value that represents their strength of feeling about a given topic

fixed choice option - includes a list of possible options and respondents are required to indiciate which one applies to them

designing interviews: must be scheduled and must be standardised to the participant to avoid interviewer bias. should be conducted in a quiet room away from others and should start with neutral questions to build up rapport and always remind them of their right to withdraw (ethics). 

writing good questions: avoid overuse of jargon (specialist terms that the average person is not aware of) avoid emotive language and leading questions (as this may change the way the participant answers) and avoid double-barrelled questions or double-negatives (so it is more clear what the interviewer is asking)

qualitative: descriptive

strengths: greater insight, higher external validity

weaknesses: harder to analyse

quantitaitve: quantifiable, usually numbers

strengths: easier to analyse and thus compare

weaknesses: loss of insight, little external validity

primary data: data that has been obtained first-hand (directly from participants in an experiment, self-report or observation) / strengths: authentic / weaknesses: costly and time-consuming

secondary data: information that has already been collected by someone else / strengths: inexpensive, acquired with minimal effort / weaknesses: may be out-dated or incomplete

meta-analysis - combining results from a number of studies that have already been conducted to give an overall view

descriptive statistics: the use of graphs, tables and statistics to identify trends and analyse sets of data

measures of central tendency: measure of the average

- mean (most sensitive as it includes all values in data set so is most representative BUT is easily distorted by extreme values as it accounts for anomolies)

- median (extreme scores do not affect it BUT it is not as sensitive)

- mode (easy to calculate BUT is not representative)

measures of dispersion: measure of variation in set of data

- range (east to work out BUT does not give fair representation of general spread)

- standard deviation: low SD suggests data is tightly clustered around the mean so participants repsonded in the same way (precise BUT can be distorted by extreme values)

peer review - 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. it establishes the validity and accuracy of any research.

aims:

- to allocate research funding

- to validate the quality and relevance of research

- to suggest amendments or improvements

evaluation: - must remain anonymous, although reviewers may use anonymity to be highly critical (may be due to competition for funding)

- problem with publication bias - journals may only want to publish 'headline grabbing' research even if the research is false or invalid

- reviewers tend to bury ground-breaking research if it contradicts their own. in order to maintain status quo, reviewers will pick research that chimes with current opinion