year 13 research methods Flashcards
why do we use a statistical test
to determine if an difference/ correlation exists in research.
significant or occurred by chance
null hypothesis
this is when the findings have occurred by chance
opposite of hypothesis
choosing a statistical test
- difference or correlation
- related vs unrelated (not correlation)
- levels of measurement
Some Women Really Can’t Measure Under Circumstances So Petty
related design
matched pairs, repeated measures
unrelated design
correlation
nominal data
a level of measurement when data is in seperate categories
ordinal data
subjective. data organised in a way but intervals inbetween each item is unequal and can be ranked against eachother
interval data
mathematically measured. the speed or time it takes someone to do something.
nominal statistical tests
unrelated- chi squared. related- sign test. association test- chi squared
ordinal statistical tests
unrelated- mann whitney. related- wilcoxon. association- spearmens rho
interval statistical tests
unrelated- unrelated t-test. related- related t-test. association- pearsons R
level of significance in psychology
0.05
chi squared degree of freedom - spread of data
(rows-1) x (collumns-1)
pearsons R degree of freedom
n-2
related t-test degree of freedom
n-1
unrelated t-test degree of freedom
na + nb - 2
type 1 errors
null hypothesis is rejected when the results were acc due to chance and the researcher is being optimistic
type 11 errors
null hypothesis is accepted when the results are actually significant and the researcher is being pessimistic.
correlation
a measure of how strongly two or more variables are related to eachother - cannot establish cause and effect
disadv of correlations
do not show causation, dame weakness as methods used to gather data, misleading, cannot establish real cause
adv of correlations
good pilot study, can research unethical variables without doing an experiment, understand relationships between variables (pos/neg)
psychological report
abstract, introduction, methods, results, discussion
features of science - empirical
data must be collected based on the theory in an experiment or observation for example
features of science - objectivity
this aims to reduce bias. may use 2 researchers and compare. lab studies help to stop bias
features of science - replicability
Popper argued a experiment or theory should be able to be replicated in order for the findings to be allowed to be generalised
festures of science - falsifiability
if you cannot test that a theory is wrong it is not scientific. e.g. freuds theories about the 3 parts of personality
features of science - paradigm shifts
complete shifts in what the theory or science believes in. e.g from Freud, to bandura, to neuroscience
features of science - a theory
a set of general laws that can explain particular behaviours etc
sign test step 1
put the data collected into a table
sign test step 2
subtract scores from condition A and B then record differences in score with +/-/= signs
sign test step 3
add up + and - and smaller number is value S
sign test step 4
compare S w critical value of that type of hypothesis and number of ppts to see if significant. if calculated value is smaller then or equal to the critical value the score is significant
