choosing statistical tst- research methods Flashcards
statistical testiong
used to determined whether difference or association correlation found is significant
outcome of this has implications for whether we accept or reject the null hypothesis
whether researcher looking for difference or correlation, what experimental design being used, level of measurement
difference or correlation
aim or purpose of investigation
obvious from wording of hypothesis
experimental design
if looking for correlation this isnt relevant
independent groups repeated measures and matched pairs
repeated measures and matched pairs-related designs
repeated measures- same pps used in all conditions of the experiment
matched pairs-pps in each condition not he same but have been matched on some variable that is important for the investigation whcihmeakes them related
independent group-unrelated pps in each condition are different, researcher chooses from related or unrelated
choosing statistical test
unrelated design and nominal data- chi sqaured
unrelated and ordinal data- mann-whitney
unrelated and interval data- unrelated t test
related and nominal- sign test
related and odinal- wilcoxon
related and interval-rleated t test
nominal data-chi squared
ordinal data-spearmans rho
interval-pearsons r
levels of measurement- nominal data
data represened in form of categories
categorical data
discrete in that one item can only appear in one of the categories
levels of measurement- ordinal data
data ordered in some way
does not have equal intervals between each unit
lacks precision becvaue based on subjective opinion rather than objective measures
unsafe data becase lacks precision
not used as part of statistical testing
raw scores converted to ranks and it is the ranks that are used in the calculation
levels of measurement- interval data
based on numerical scales include units of equal precid=sely defined size
more detail preserved
most precise and sophisticated form
necessary criterion for se of parametric tests
relation to measures of central tendancy
nominal-mode
ordinal- median and range
interval- mean and standard deviation