OCR a-level psychology research methods flashcards
what is the purpose of peer review
to ensure that only high quality psychological research is published into the public domain
step one of peer review
psychologist sends the editor of a magazine or journal their research
step two of peer review
the editor examines the topic of the research and send it to experts on the field
step three of peer review
the reviewers assess it for the editor
step four of peer review
reviewers send the research back to the editor with comments and a recommendation of what to do with it
step five of peer review
the editor decides for themself if the research should be accepted for publication, sent back for revisions, or rejected
why does peer review check about a piece of research
it’s validity
limitation of peer review
failure to detect fraudulent data, rejection on the basis it goes against the status quo, bias, positive results being more favoured than negative or null ones
purpose of a title in a report
inform readers what the report is about
purpose of the abstract in a report
a brief summary of the paper
purpose of an introduction in a report
provide background details of the topic and study
purpose of a method in the study
tells readers how the research was done
purpose of results in a study
summarise the findings of the research
purpose of a discussion in a report
discuss the findings and their implications
purpose of references in a report
informs readers of where the researchers found their sources
purpose of appendix
to inform readers of details not mentioned in the report
naughty words
difference, cause and effect, IV, DV, effect, conditions, groups, between
what is a positive correlation
a relationship between variables where, as one increases, the other also increases
what is a negative correlation
a relationship between variables where, as one decreases, the other increases
association does not
mean causation
what is no correlation
when variables are considered uncorrelated
what does correlations refer to
a measure of how strongly two or more variables are related to each other
what level of data does correlational analysis need
at least ordinal data
structure of a null hypothesis
there will be no significant correlation between [variable one] (measured by) and [variable two] (measured by). any relationship found will be due to chance.
structure of a one-tailed hypothesis
there will be a significant (pos/neg) correlation between [variable 1] (measured by) and [variable 2] (measured by).
structure of a two-tailed hypothesis
there will be a significant correlation between [variable 1] (measured by) and [variable 2] (measured by).
what level of data must there be for a correlational analysis
at least ordinal data
between what numbers is there a strong positive correlation
0.5 and 1
between what numbers is there a strong negative correlation
-0.5 and -1
between what numbers is there a weak positive correlation
0 and 0.5
between what numbers is there a weak negative correlation
0 and -0.5
what does 1 mean as a correlation coefficient
perfect positive correlation
what does -1 mean as a correlation coeffecient
perfect negative correlation
what does 0 mean as a correlation coefficient
there is no correlation
strengths of a correlational study
- identify relationships between variables that would be impractical or unethical to manipulate in an experiment
- correlations are a good starting point to suggest ideas for future research
weaknesses of a correlational study
- cause and effect cannot be established as correlations can only suggest a relationship between two variables, so further research would be needed
- an unknown third variable may have caused the link, therefore correlations may lack validity
what kind of distribution curve is this
normal distribution curve
what kind of skewed distribution is this
left skew/negative distribution
what kind of skewed distribution is this
right skew/positive distribution
what is the mode
the most frequent score
what is the median
the middle point in the data
what is the mean
adding all the data and dividing by how many scores there are
what does probability mean in psychology
the likelihood that the pattern in the data could be due to chance
by assessing the probability…
…we can determine the significance of the results
high significance
low probability
high probability
low significance
type 1 error
false positive
what hypothesis is rejected but correct in a type 1 error
null hypothesis
what hypothesis is rejected but correct in a type 2 error
directional hypothesis
type 1 error
false positive
type 2 error
false negative
what error happens when the probability is too strict
type 2 error
what error happens when the probability is too lenient
type 1 error
68% of participants
percentage of participants that are within 1 standard deviation of the mean
95% of participants
percentage of participants within 2 standard deviations of the mean
99.7% of participants
percentage of participants within 3 standard deviations of the mean
two requirements of parametric tests
normal distribution, interval data
what stats test is for unrelated data at a nominal level
chi squared
what stats test is for related data at a nominal level
binominal sign
what stats test is for unrelated data at an at least ordinal level
mann whitney u
what stats test is for related data at an at least ordinal level
wilcoxon signed
what stats test is for correlations
spearman’s rho
if the tests name has an R in it then the observed value must be…
greater than or equal to the critical level in order to be significant
if the tests name doesn’t have an R in it then the observered value must be...
less than or equal to the critical level in order to be significant
what is a lab experiment
an experiment conducted in a highly controlled and artificial environment. the researcher manipulates the IV.
what is a field experiment
an experiment conducted in a natural environment. the researcher manipulates the IV.
what is a quasi experiment
an experiment with a naturally occurring IV. the research cannot manipulate the IV.
strengths of a laboratory experiment
high levels of control, no possible EVs, cause and effect can be established
high levels of control, replicable, reliable
weaknesses of a laboratory experiment
artificial setting, low ecological validity, cannot be generalised
may not act naturally, demand characteristics, low internal validity
strengths of field experiments
high ecological validity, low chance of demand characteristics
weaknesses of field experiments
low levels of control, ev’s, cause and effect can’t be established
difficult to replicate, decreases reliability
strengths of quasi experiments
can use an IV that would be unethical/impossible to replicate, practical application
often done in a labratory setting, high control
weaknesses of a quasi experiment
participants can only belong to one condition, individual differences and less valid
some IVs are not frequently occurring
examples of experimental methods
lab, field, quasi
examples of experimental designs
independent, repeated, matched pairs
what is independant measures
participants are randomly allocated one condition
what is repeated measures
participants take part in all conditions
what is matched pairs
researchers make sure each participant has a match in terms of key characteristics in each condition
strengths of independent measures
no risk of order effects
weaknesses of independant measures
high risk of individual differences
ineffective as each participant is only used once
strengths of repeated measures
no risk of individual differences
efficient use of participants as they are used more than once
weaknesses of repeated measures
high risk of order effects
more prone to demand characteristics
strengths of matched pairs
no risk of order effects
reduces the effect of individual differences
weaknesses of matched pairs
very difficult to successfully put in place
how to deal with order effects
counter-balancing
what is a situational variable
anything in the environment that can affect participants behaviour
examples of situational variables
time of day, sound level, temperature
how to control situational variables
keep the environment as similar as possible for all participants
what is a participant variable
differences between participants that aren’t accounted for in the IV
examples of participant variables
gender, age, cultural background, mood, intellect, anxiety, amount of siblings etcetcetc
how to control participant variables
a large sample size limits participant variables, random allocation to groups, matched pairs design
what are experimenter variables
the experimenter may unconsciously convey to participants how they should behave (experimenter/researcher bias)
examples of experimenter variables
demand characteristics → please you/screw you effect, social desirability, expectation effects
how to deal with experimenter variables
single-blind procedure, double-blind procedure, placebo conditions, standardised instructions
what is a single-blind trial
where the participants do not know the aim of the study/what condition they are in to reduce the effect of demand characteristics
what is a double-blind trial
where neither the participants or researchers know the true aims/conditions of the study, to reduce the effect of researcher bias and demand characteristics
what are placebo conditions
where participants believe they are recieving something but they are not, to lessen the effect of social desireability
what are standardised instructions
every participant is given the same amount of information so there is a lesser chance the investigator can convey experectations
who does controls apply to
all conditions
who does standardisation apply to
just one condition