RESEARCH METHODS Flashcards
Aim
General statements that describe the purpose of investigation
Hypotheses
Desrcibes relationship between variables at start of study either directional or non directional
Directional hypothesis
States direction of the difference or relationship inc words such as more, faster and higher
Non directional hypothesis
Doesnt state the direction of the difference or relationship words such as less, lower and slower.
IV
Changed
DV
Measured
Operationalised
Clearly defined variables in terms of how they can be measured
Extraneous variables
Any variable that affects the DV if not controlled do not vary systematically with the IV
Confounding variable
a kind of extraneous varibale but changes systematically with the IV
Investigator effects
Effect of behaviour on the research outcome l.
Randomisation
Minimises effect of extraneous/confounding variables on the outcome. Use of chance methods to reduce the researchers unconscious biases.
Standardisation
Using the same formalised procedures
Experimental design
Different ways to organise ps to conditions
Independent group design
Ps allocated to different groups where each group represents one condition
Repeatedmeasures
All ps take part in all conditions
Matched pairs
Ps are matched on one variable that may affect the DV and each p does one condition
independent groups AO3
- participant variables
- order effects reduced
- more money+time
- use random allocation to reduce participant variables
Random allocation
Ps randomly allocated to the different conditions
Repeated measures AO3
- order effects
- less money+time
- demand characteristics
- counterbalance to reduce order effects
Counterbalancing
Half the ps take part in condition A then B and other half do the other way round reduces order effects
Matched pairs AO3
- time consuming
- not matched exactky
- order wffects and demand characteristics less of problem
Quasi experiment
Iv has not been determined by anyone
Natural experimebt
The change in iv is not brought about by researcher.
Random sample
All members have equal chance of being selected lottery method
Systematic sample
Where every nth memeber of the target population is selected
Opportunity sample
Select anyone willing and available
Volunteer sample
Select themselves to be part of sample
Random sample AO3
- unbiased so increased internsl validity
- time consuming
- may be unrepresentative
Systematic sample AO3
-  Objective
- time consuming
- ps may refuse
- researcher has no influence
Stratified sample AO3
- representative
- generaliseable
Stratifued sample
Composition of sample reglects proportiond of people in population
Opportunity sample AO3
- less costly
- unrepresentative
- researcher bias
Volunteer sample AO3
- easy
- less time consuming
- volunteer bias
Types of ethical issues
Informed consent,deception,protection from harm, privacy
Pilot studies
Small scale investigation to check the procedures.
Single blind proceeure
When ps are aware of the aim of the research
Double blind procedure
Neither ps or researcger is aware of aims of the study
Naturalistic observation
Watching and recirding behaviour in a natural environment
Controlled observation
Watching and recording behaviour in a structured environment
Covert observation
Ps behaviour is watched and recirded without their knowledge
Overt obversation
Ps behaviour is watched and recorded with their knowledge
Participant observation
Researcher become a member of gorup whose behaviour they are recording
Non participation observation
Researcher remains outside grouo whose behaviour they are recording
AO3 of Observations
Observation bias
Naturalistic observations AO3
- generalised
- lack of control so cant replicate
Controlled observations AO3
- extraneous variables controlled
- not applied to every day life
Covert observation AO3
- less demand characteristics
- consent issues
- internal validiry
Participant observation AO3
- increased insight increasing external validity
- lose objectivity
Non participant observation AO3
- objective
- lose valuable insight
Behavioural categories (observations)
Target behaviour is broken down into components that are measurable
Event sampling observations
A target behaviour is recorded this event every time it occurs
Time sampling observations
A target individual or group is established and the baby behaviour is recorded in a fixed timeframe
 correlation
Investigate an association between two variables
Measures of central tendency
Mean, median and mode
Measures of dispersion
Range and standard deviation
Standard deviation
How mych each score deviates from the mean
case studies
involve analysis of unusual individuals or events. Qualititative data. Longitudinal study.
content analysis
research technique that enables the indirect study of behaviour by examining communications
content analysis aim
summarise and describe this communication in a systematic way so overall conclusions can be drawn.
coding in content analysis
stage of content analysis in which the communication to be studied is analysed by identifying each instance of the chosen categories
thematic analysis
an inductive and qualitative approach to analysis that involves identifying implicit or explicit ideas within the data. Themes often emerge once the data has been coded.
case studies strengths
- rich detailed info
- generate hy[otheses for future study
case studies weaknesses
- generalisation
- subjective selection and interpretation of researcher
- low validity
content analysis strengths
- little ethical issues as it already exists in public domain
- external validity
- flexible as produces qualitative and quantitative date depending on aim
content analysis weaknesses
- analysed outside of context where it occurred
- researcher may attribute opinions and motivations to the writer
- lack of objectivity
- thematic analysis is more descriptive
reliability
how consistent a measuring device is
ways of testing reliability
- test retest
- inter-observer reliability
- measuring reliability
test retest
doing same tests or questionnaire to the same person on different occasion.
test retest procedure
time between test and retest to remove memory variable. The two setoff scores would be correlated. If correlation is significant then reliability is good
inter observer reliability
subjectivity and bias issues so observers check they apply behavioural categories the same and compare results.
measuring reliability
correlational analysis correlation coefficient should exceed +80 for reliability
validity
whether a psychological test, observation or experiment produces a result that’s legitmate
internal vailidty
whether the effects observed in an experiment are due to manipulation of independent variable and not another factor.
external validity
generalising to other settings, populations and eras
face validity
measure is scrutinised to determine whether it appears to measure what its suppose to measure
concurrent validity
extent to which a psychological measure relates to existing similar measure
temporal validity
findings from research study can be generalised to other historical time and eras
when to use a directional hypothesis
when previous research suggest a specific outcome
when to use a non directional hypothesis
when there are no previous research or it contradicts each other
what are correlations plotted on
scattergrams with one co variable on x axis and other on y axis
qualitative data
data expressed as words
quantitative date
date that can be counted usually numbers
qualitative dataAO3
- rich in detail
- broader in scope
- external validity
- hard to analyse
- subjective interpretations
quantitative data AO3
- easy to analyse
- make comparisons easier
- more objective
- less bias
- less detail and narrow
ways of testing reliability
- test retest
- inter-observed reliability
validity types
- face
- concurrent
- ecological
- temporal
- internal
- external
concurrent validity
psychological measure relates to an existing similar measure
ecological validity
findings can be generalised to other settings
temporal validity
findings can be generalised to other eras
3 things that determine statistical testing
- correlation/experiment
- experimental design
- level of measurement
nominal related
sign test
nominal unrelated
chi squared
ordinal related
wilcoxen
ordinal unrelated
mann whitney U
internal related
related t test
interval un related
unrelated t test
correlation nominal
chi squared
correlation ordinal
spearman
correlation internal
Pearsons
reporting psycholigical investigations
abstract, introduction, method, results, discussion, referencing
the abstract includes
the aims, hypothesis, method, results and conclusions
method should include
design, sample, apparatus, procedure, ethics
paradigm
a set of shared assumptions and agreed methods
paradigm shift
result of scientific revolution
empirical
based on evidence through observation and experience
falsifiability
theory cannot be scientific unless it can be proved untrue
