Psych Stats Final

type 1 error

occurs when a researcher rejects a null hypothesis that is actually true

type 2 error

occurs when a researcher fails to reject a null hypothesis that should be rejected

alpha levels

the probability of obtaining data in the critical region if hypothesis is true

fails to detect it

a type 2 error means that an effect exists, but the hypothesis test…

effect exists

a type 1 error means that a treatment effect does NOT exist, but the hypothesis says the…

true

a between-subjects design means there are separate groups of participanrs for each group of scores

independent-measures design

between-subjects design is also known as

repeated-measures design

a _, also known as a within-sibjects design, collects several groups of scores from the same group of participants

correlation

_ is a statistcal technique that is used to measure and descrive the relationship between two variable

2

in a correlation study, there is 1 group of participants and _ variables measured

person r test

measures the degree and direction of the linear relationship between two continuous variables

positive

in a _ correlation, the two variables change in the same direction (as X increases, Y increases)

negative

in a _ correlation, the two variables tend to go in opposite directions (as X increases, Y decreases)

ANOVA

_ is a hypothesis-testing procedure used to evaluate mean differences between two or more treatments

true

post hoc tests are completed after an ANOVA to determine exactly which mean differences are significant and which aren’t

increases

ANOVAs are used when comparing 2+ groups as opposed to several t-tests, using several t-tests _ risk of type 1 error

nonparametric

_ tests use nominal or ordinal scales that do not produce numerical values

main effect

the _ refers to the mean differences among the levels of one factor

interaction

an _ between two factors occurs when mean differences between treatments are different than the overall main effects

false

in a two-factor ANOVA, a significant interaxtion exists when effects of one factor are NOT impacted by the other factor

nominal scale data

parametric tests can not use…

ratio scale data

nonparametric tests can not use…

sampling distribution of differences

distriubtion of all possible differences between two sample means

estimated standard error of difference

standard deviation of sampling distribution of differences

independent samples t-test

tests for statistically significant difference between a pair od sample means from a between-subjects research design

within-subjects research design

comparison of same sample at two different time periods

post hoc tests

determines exactly which mean differences are statistically significant, difference between means must be greater than or equal to HSD to be statistically significant

two-factor ANOVA

comparison of two or more means

between-treatment variance

variability in scores between treatment conditions

within-treatment variance

variability in scores within each treatment condition

correlation

statistical technique to measure and describe the relationship between two variables

restricted range

can underestimate true population correlation

outliers

can change strength and direction of a correlation

sperman correlation

relationship between two variables on an ordinal scale

point-biserial correlation

relationship between one continuous variable and one dichotomous variable

phi-coefficient

relationship between two dichotomous variables

regression

statistical technique for finding the “best-fitting” straight line for a set of data

parametric tests

test hypothesis about population parameters

nonparametric tests

available when research data violate requirements/assumptions for parametric tests

chi-square test for goodness of fit

tests proportions or relative frequencies for a distribution

chi0square test for independence

evaulates relationship between two variables

mann-whitney u-test

alternative to the independent samples t-test, compares scores between two different groups

wilcoxon signed-ranks test

alternative to the dependent samples t-test, compares scores from a within-subjects research design

kruskal-wallis test

alternative to the one-way ANOVA, cmpares scores between two or more different groups

friedman test

alternative to the repeated measures ANOVA, compares scores from within-subjects research design