FINAL - data analysis: inferential Flashcards
inferential statistics
test how likely it is that the study findings represent the population
sampling error
errors that can happen, even if random selection occured
alpha .05
probability of error is 5/100
95% confidence
alpha .01
probability of error is 1/100
99%
alpha .001
probability of error is 1/1000
99.9%
p value
probability of rejecting the null hypothesis
what is the likelihood that your results are due to sampling error?
used in determining if results are considered “statistically significant”
which would be more likely to be used with a directional hypothesis?
one tailed test
type II error
failure to reject the null hypothesis
often caused by the study not having enough power to find statistically significant results
mainly controlled through planning/follow through of study
correlation coefficient
r
parametric
when determining relationships, need to look at:
strength of relationship
direction of relationship
statistical significance of relationship
you cannot tell cause/effect based on
correlations
WEAK RELATIONSHIP
0.3-0.5
MODERATE RELATIONSHIP
0.51-0.85
STRONG RELATIONSHIP
0.86-0.95
positive r
both variables change in the same direction
negative r
variables change in different directions
t test
analyzing differences between 2 groups/conditions
ANOVA
analyzing differences for MORE THAN 2 groups/conditions
parametric tests
independent tests
paired t test
z test
nonparametric tests
mann-whitney U
sign test
wilcoxin matched-pairs signed-rank test
one way designs
one way analysis of variance (ANOVA)
repeated measures ANOVA
factorial designs
two way analysis of variance
mixed model analysis of variance