research methods part 4/5 Flashcards
what does science allow us to do ?
make predictions and control the world
what are the 4 scientific methods that we can use in a study ?
- empirical methods & objectivity
- replicability and falsifiability
-theory constructions & hypothesis testing - paradigms and paradigm shift
what is the aim with empirical methods ?
to gain facts
why is empirical methods important ?
because people can make ‘claims’ on anything but the only way we can verify it to be true is through direct testing
what produces empirical evidence ?
direct testing
what is objectivity ?
basing findings on facts rather then opinions
does something to be scientific need to be objective or subjective ?
objective
what is replicability ?
where we have the ability to repeat a study and test the findings aren’t just a ‘fluke’
- as if it isn’t a fluke then it is reliable
what is necessary for re-testing to happen similarly ?
clear standardised procedures
what is necessary for a scientific theory to be trusted ?
must be repeatable over a number of different context and circumstances
who created the theory that for a study to be truly scientific it needs to be falsifiability ?
karl popper (1934)
what does falsifiable mean ?
for a theory to be proven wrong
why is falsification good for a theory ?
as the more a hypothesis is able to stand up against this process the more likely it is to be true
what is a theory ?
is a set of general laws or rules to explain events or behaviours
what might a theory but put foward for ?
a particular behaviour
what does theory construction include ?
induction
deduction
what does the induction in theory construction resemeble ?
involves the reasoning from the particular to the general
what does the deduction theory resemble in theory theory construction ?
involves reasoning from the general to the particular
what does the theory construction form the basis of ?
of a hypothesis
what is a paradigm ?
a particular set of assumptions or a generally accepted way of thinking wishing a subject or discipline
what did Kuhn (1962) suggest about psychology and social sciences ?
that psychology and other social sciences are not scientific as they do not posse a single paradigm as they have a number of competing assumptions
how often did Kuhn suggest a paradigm shift happens ?
ever so often
what is a paradigm shift ?
where a new way of thinking takes over
when would a paradigm shift happen ?
once enough evidence emerges to challenge the current paradigm
what is a case study ?
an in-depth investigation of a single person, group or event, community
how is case studies typically gathered ?
gathering data in a variety of ways/ sources from different methods
what are the 3 key features of a case study ?
- qualitative data
- quantitative data
- likely to be conducted over a long period of time
how does the researcher gather quantitative data ?
potential experimental testing
what does the researcher use qualitative data for ?
to construct a case history of the subject
what is 2 examples of a caste study ?
- Phineas Cage
- Genie Wiley
what are 2 strengths of case studies ?
- rich, detailed information that focuses on 1 person
- useful at contributing to our understanding of ‘normal’ behaviour
what are 2 weaknesses to case studies ?
- case studies only focus on one person or very small numbers
- they rely on personal accounts of behaviours and experiences
what is content analysis ?
a type of analysis of any from of communication that people produce
why for content analysis are people studies are indirectly ?
because its their communication that’s studies rather than the person themselves
what type of data is used for content analysis ?
- qualitative
- quantitative
- or both
what are examples of communication that can be studies for content analysis ?
- emails
- text messages
- tv programmes
- films
- newspapers
- magazines
- presentations
- interviews
- conversations
what is the aim of content analysis ?
summarise and describe the information in a systematic way so we can draw conclusion
what process is content analysis similar too ?
observation studies
- but observe books instead of people
what does the researcher create in stage 1 ?
quantitative data
what is the first stage in conducting content analysis include ?
coding
what does the researcher do in the first stage of content analysis ?
researcher creates behavioural categories
what does the researcher do with the data after from stage 1 ?
the researcher categories the data into meaningful units
what does the second stage of content analysis include ?
thematic analysis
what 2 things need to be taken into consideration during content analysis ?
- the sampling methods
- how the data should be recorded
what is the idea of the second stage of content analysis ?
to get some kind of order from the data
- it allows you to summarise and reduce the data
what does the second stage of content analysis create ?
qualitative data
what can the researcher do in content analysis once the data has been coded ?
they can then identify themes
what is a theme in content analysis ?
a theme is something that keep ‘cropping up’ and can only really be done once the data has been coded
- allows you to draw conclusions
what are the 2 strengths to content analysis ?
- can get around ethical issues that may exists with other methods
- practical application
-material already exists so can be replicable
what are 2 weakness for content analysis ?
- people are studied indirectly
- also an issue with culture bias
what does validity mean ?
how accurately a method measures what it is intended to measure
what are the 2 types of validity ?
- internal
- external
what does internal validity relate too ?
whether the results can be attributed to the other effects of the manipulated variable alone rather than to any other ‘nuisance’ variables
- IV should be the only thing that affects the DV so we can establish the cause and effect
what is internal validity affected by ?
- whether or not the measures taken or tasks used were actually testing what they were supposed to text
what can allow a study to have low internal validity ?
- demand characteristics
- order effects
- bias
- confounding variables
what are 2 types of external validity ?
- ecological
- temporal
what is ecological validity ?
can the findings be generalised beyond the setting in which the study was performed
what is mundane realism ?
is the study like ‘real life’
what is temporal validity ?
refers to how relevant the time period is in affecting the findings
what are 2 ways to assess validity ?
- face validity
- concurrent validity
what is face validity ?
looking at a measure/ test/ scale and deciding ‘on the face of it’ if it is valid
- so ask an expert to look over it
- degree to which a measure appears to be measuring what it is intended to measure
what is concurrent validity ?
measures the extent to which a measurement is confirmed by a related measurement
what are 3 ways to improve an experiments validity ?
- use a control group
- use standardised procedures
- use double blind procedures
what are 2 ways to improve questionnaires validity ?
- use a lie scale
- maintain anonymity
what is a way to improve observations validity ?
- use covert, naturalistic observations
what are 2 ways to improve qualitative methods validity ?
- use direct quotes
- use triangulation
what does reliability mean ?
the degree to which research methods can produce consistent and stable results
what can you describe a measure that can be repeated ?
reliable
what are the 2 ways to asses reliability ?
- test re-test
- inter-observer reliability
what is the test re-test method ?
they do it once then tested again on a different occasion
- some time will be in-between but not too long to avoid changes in attitudes and behaviours
what correlation coefficient is needed to say the tool has test re-test and inter-observer reliability ?
0.8 or higher
what is inter-observer reliability ?
if there is more than 1 observer you need to make sure they’re interpretating the events the same
- as need to avoid/ reduce subject interpretations
how do you make sure that experiments are more likely to be reliable ?
- use lab experiments as the researcher can control many variables that may affect the outcome
what can help make questionnaires more reliable ?
if the test re-test correlation is lower then 0.8 than some questions may need to be change or all
what can help make observations more reliable ?
by making sure behavioural categories have been operationalized
what can help make qualitative data more reliable ?
- use the same trained interviewer and use structured interviews to avoid leading questions and have more control on psychologist behaviour
what is another name for the alternative hypothesis ?
directional or non-directional hypothesis
do we want to falsify the null hypothesis or not ?
yes
what do we use statistical tests for ?
to work out whether a difference is due to chance or a real different
what does chance refer too ?
something with no cause it just happens
what number is sued to determine if the results are statistically significant or not ?
calculated value
what is a large impact as to why the psychologist can never be 100% certain ?
due to individual differences
what does ‘significant at p<0.05’ ?
what are the 2 errors that can be made ?
- type 1
- type 2
what is a type 1 error ?
when it is too lenient
- the null hypothesis is rejected and the alternative is accepted when it should’ve been the other way round
- eg) using 0.1 chance it is a fluke
what is a type 2 error ?
when it is too strict
- the alternative hypothesis is rejected and the null is accepted but it should’ve been the other way round
- eg) using a 1% chance the result is a fluke
what are the 3 D’s in deciding which statistical test to use ?
- design
- difference
- data
what are the 2 types of test to look out for in statistical tests ?
- difference
- correlation
what are the 3 designs to look out for in a statistical test ?
- independent group
- repeated measure
- matched pairs
what are the 2 related designs for a stats test ?
- matched pairs
- repeated measure
what is the unrelated design for a stats test ?
- independent group
what are the 3 ways data can be measured ?
- nominal
- ordinal
- interval
what is nominal data ?
data in categories
what is ordinal data ?
data that can ordered in some way
- doesn’t need to have equal intervals
- lack precision as based on opinion
- not measuring something real as an opinion
what is interval data ?
- based on ‘real’ numerical scale with units of measurement that is clearly defined
- eg) temp, time, height, weight
what data is a parametric test for ?
interval
when is parametric tests used ?
when the data is from a population which can be assumed to be normally distributed data which has a ‘homogeneity of variance’
what does ‘homogeneity of variance’ mean ?
similar dispersion
what are the 3 D’s for the Chi squared for difference ?
- difference
- unrelated
- nominal
what are the 3 D’s for the Mann-Whitney ?
- difference
- unrelated
- ordinal
what are the 3 D’s for the Unrelated t test ?
- difference
- unrelated
- interval
what are the 3 D’s for the Related t test ?
- difference
- related
- interval
what are the 3 D’s for the Sign test ?
- difference
- related
- nominal
what are the 3 D’s for the Wilcoxon ?
- difference
- related
- ordinal
what are the 2 D’s for the Chi- squared for a correlation ?
- correlation
- nominal
what are the 2 D’s for the Spearman’s Rho ?
- correlation
- ordinal
what are the 2 D’s for the Pearson’s r ?
- correlation
- interval
what statistical test mean the calculated value need to be equal to or greater than the critical value ?
- spearman’s Rho
- chi squared
- pearson’s R
- unrelated tests
- related tests
what statistical test mean the calculated value need to be equal to or less than the critical value ?
- Mann Whitney
- Wilcoxon
- Sign test
what is the rule of ‘R’ ?
if there is an R in the name then the calculated value needs to be greater than the critical value
