psych stats final

when do you use a one-samples t-test?

- data for a variable of a single group and comparing data to the population

-data for a variable of a single group and comparing result to a known value

-data for a variable of a single group and comparing result to untreated population

a one-sample t-test does NOT

compare data between experimental and control groups

example of a one-sample t-test

comparing local schools IQ scores to the national average of 100

when do you use a t-statistics?

when the population standard deviation is unknown

independent-measures t-test/ independent samples/ independent groups/ between-subjects design

data from two completely different independent participant groups (experimental and control)

-based on two means/two standard deviations/ two degrees of freedom —> pool them to calculate standard error

-compare two seperate samples with unknown population parameters

chapter 9

independent measures and repated measures

within-subjects design/ repeated-measures design

data from same participants (one group)

-reaction time on words v. colors

chapter 10

independent-measures t-test

chapter 11

repeated measures design and difference of means

paired samples t-test

examines two groups who do opposite order of two tasks

given raw data of differences and need to do calculations

analyzes data from two related samples, such as pre-test/post-test scores or matched pairs, where each subject (e.g., person, machine) is measured twice

ANOVA means

analysis of variance

analysis of variance

used to evaluate mean differences between 3 or more treatments/populations

uses sample data as the basis for drawing general conclusions about populations

ANOVAs advantage to that of a t-test

can be used to compare 2 or more treatments at the same time

ANOVA examples

-measuring the effect of 2 different drugs and a placebo

-measuring the effectiveness of 3 study methods

factor (ANOVA)

independent/quasi-independent variable that indicates the groups being compared

levels of the factor

individual conditions/values that make up a factor

factorial design

a study that combines 2 or more factors

ways H0 can be wrong

-all means are different from every other mean

-some means are not different from others, but other means do differ from some means

why ANOVA and not t-tests

Experiments often require multiple hypothesis tests (each type 1 error/false positive)

Type 1 error for set of tests accumulates alpha (0.05 per test - # of tests)

ANOVA evaluates all mean differences at the same time/test

and avoids the problem of inflated experimentalwise alpha

you can not compute ________ between 2+ samples

sample mean difference

f ratio- variance

instead of sample mean difference

between-treatments variance

variability results from general diferences betwen the treatment conditions (numerator)

within-treatments variance

variability within each sample. individual scores are not the same within each sample (denominator)

(ANOVA) k

number of treatment conditions

(ANOVA) n1 and n2

number of scores in each treatment

(ANOVA) N

total number of scores

(ANOVA) T

sum of scores for each treatment

(ANOVA) G= sum of T

grand total