Stats 1 Flashcards
what is the standard deviation
mesure variabilty in data
what is z score
used for comparing a range of measurements
express how many sd units away from the mean a point in normal curve is
what is central limit theorem
distribution of all sample means will approach normal distribution
whilst individual sample means may deviate from the population mean,the mean of all sample means will equal the population mean
as the sample size increases, SD of sampling distribution decreases
what is the standard error of the mean
a statistic that represents the standard deviation of the sampling distribution
what does standard error represent
how confident we can be that our sample mean represents the population mean
what is population mean
mean of the sample means
what is type I error
rejecting null hypothesis when its true
what is type II error
accepting null hypothesis when its false
how does a small sample differ from normal sample on graph
pointier around mean
fatter around extreme
what determines df
shap of graph
how do we measure homogeneity of variance
examine sds
levenes test
what is the non parametric alternative for single sample t test and independent samples t test
wilcoxon signed ranks
mann-whitney u-test
what does a conservative alpha rate do
raises the chance of type II errror
how is familywise error corrected
using bonferroni method
what are two problems with multiple comparisons
computationally inefficient
increases overall probability of type 1 error
what is familywise error
the probability of making one or more false discoveries, or type I errors when performing multiple hypotheses tests
what is ANOVA
a stat test designed for factors with more than two levels that is computationally efficient
does not compromise type I error
what is between groups variance
the deviations of the group means from grand overall mean
describe the trend of between groups variance
the greater the group means differ from the grand mean, the bigger this will be
what is within groups(error)variance
the deviation within each group from each group mean
describe the trend of within groups (error) variance
the more scores within each vary from each other, the bigger this will be
what should between groups variance consist of
any treatment or condition effect plus measurement error
what should within groups (error) variance consist of
should consist only of measurement error
what does between groups vdf relate to
number of conditions
what does within groups(error) df relate to
participants per condition
what does the shape of f distribution depend on
2 dfs
what is degrees of freedom
the number of values in the calculation of each mean that free to vary
how do we calculate between groups df
nc-1
how do we calculare within groups (error) df for nk equal and nk varies
nc(nk-1)
N-nc
how is f distribution determines
by its two degrees of freedom
when is an f value significant
if it exceeds critical value(the value below which 95% of distribution lies)
how do we determine the distribution
examine histograms
what is between subjects ANOVA
compare different groups to determine whether any of the groups are significantly different from each other
ANOVA equivalent of independent samples t-test
what is nonparametric alternative for between subjects ANOVA
kruskal-wallis test
how do we calculate between subjects ANOVA
sums of squares
calculate dfs
mean squares
test statistic
what is within subjects ANOVA
compares different observations from the same group to determine whether any of the sets of observations are significantly different
equivalent to paired samples t test
what is non-parametric alternative to within subjects ANOVA
friedman test
what is an additional assumption for within subjects ANOVA
sphericity
what is sphericity
the variances of the differences between conditions should be equal
how is sphericity measured
mauchlys sphericity test
what are pairwise comparisons
tell us there is a difference between some of the conditions
how do we follow up for specific hypotheses
planned or priori contrasts
how do we follow up for no specific hypotheses
post hoc tests
what is planned contrasts
compare only those conditions which you have hypotheses using t tests
how does bonferroni correction work
divide the alpha by the number of comparisons you are making
what are post hoc tests
compare all cases using an accepted post hoc using tukeys HSD
describe post hoc tests and family wise error
do best to control it without making alpha so conservative that significance is missed
compare bonferroni and tukeys
b = most conservative
what are general naming conventions
a design with x factors is an x-way design
what are specific naming conventions
each level in IV listed 3x3x4
why are complex designs unusual
complicated to interpret
between subjects; require large number of participants
within subjects; take too long per participant
when are factorial designs needed
more than one IV independently affects DV
when more than one IV combine to affect a DV
how are IV independently affecting DV determines
looking at ANOVA main effects
how are IV combined to affect DV determined
looking at ANOVA interaction effects
what are the main effects
most straightforward results within a factorial design
what do main effects represent
how one iV irrespective to the others affects the DV
what are main effects based on
summary data at the level of individual IVS(margiginal means)
how do you read a 2x2 chart to determine main effects are present
look at marginal means
if they are identical, no main effecr
what is problem with main effects
can be misleading
what are interactions
more complex results within a factorial design
what do interactions represent
how two or more IVs combine to affect DV
how do we know if there is an interaction
lines are crooked
what happens if we have an interaction
the effects of one IV are dependent on the state of the other IV
how do we measure covariance
sphericity
how many f values do factorial anovas have
3
how do you calculate the mean in excel
=AVERAGE()
how do you calculate SD in excel
=STDEV()
how do you calculate variance in excel
=POWER(A,2)
how do you compute descriptive statistics in SPSS
analyse
descriptive statistics
descriptives
how do you calculate a t test in excel
+TTEST(array1, array2, tails, type)
what is assumed if levenes is significant
unequal variances
how do we count the number of boxes in excel
=COUNT
how do we calculate the f value in excel
=F.INV
what do we calculate for a between subjects ANOVA in SPSS
univariate
what is the kruskal wallis test
compares the sums of groups ranks
how do we know if kruskal wallis test is significant
one groups summed ranks value is much higher than the other groups
what is friedman test
compares the sums of treatment conditions ranks
what do between subjects data always require in SPSS
coding variables
what do we always want from levenes and mauchlys tests if we are to assume with our tests
not significant
how do we report a significance of .000
.001
how do we find means for factorial ANOVA
looking at marginal means table
what does bonferroni correction do for familywise error
keeps it at 0.05 level of significance
which row in an SPSS table to do we look at if a mauchlys test is not significant or not available
sphericity assumed
what row in a SPSS table do we look at if mauchlys is significatn
greenhouse geisser
