EXST 2201 EXAM #3 QUESTIONS AND ANSWERS WITH
COMPLETE SOLUTIONS VERIFIED LATEST UPDATE
dependent samples
when the individuals selected for one sample do influence which individuals are in the
second
-AKA: matched-pair sample; paired samples
independent samples
when the individuals selected for one sample do NOT influence which individuals are in
the second sample
assumptions for dependent samples
1. random samples of matched pairs
2. sample average of the difference data is normally distributed
3. population standard deviation is NOT known
the mean difference between the paired samples is zero(the samples are the
same)
what is the null hypothesis for paired-sample hypothesis test?
equal population standard deviations
-best estimate of population standard deviation comes from pooling together the two
sample standard deviations
-a t-statistic using the pooled standard deviation follows a t-distribution with (n-2)
degrees of freedom
unequal population standard deviations
, -no exact method of inference exists because it is impossible to determine the exact
degree of freedom
-two alternative approximations:
1. Welch's Approximation of Degrees of Freedom
2. Satterwaite's Approximation of Degrees of Freedom
Welch's Approximation of Degrees of Freedom
take the smaller number of observations (n1 or n2), then subtract to determine the
degrees of freedom for the t-test
-best method by hand
Satterwaite's Approximation of Degrees of Freedom
a formula usually used by the computer
**MORE ACCURATE calculation
independent samples
Sample 1: the weights of 19 newborn females
sample 2: the weights of 19 newborn males
independent samples
sample 1: the score of 13 students who took the ACT
sample 2: the scores of 13 other students who took the SAT
paired or dependent samples
sample 1: the pre-training weight of 17 athletes
sample 2: the post training weights of the same 17 athletes
ANOVA
COMPLETE SOLUTIONS VERIFIED LATEST UPDATE
dependent samples
when the individuals selected for one sample do influence which individuals are in the
second
-AKA: matched-pair sample; paired samples
independent samples
when the individuals selected for one sample do NOT influence which individuals are in
the second sample
assumptions for dependent samples
1. random samples of matched pairs
2. sample average of the difference data is normally distributed
3. population standard deviation is NOT known
the mean difference between the paired samples is zero(the samples are the
same)
what is the null hypothesis for paired-sample hypothesis test?
equal population standard deviations
-best estimate of population standard deviation comes from pooling together the two
sample standard deviations
-a t-statistic using the pooled standard deviation follows a t-distribution with (n-2)
degrees of freedom
unequal population standard deviations
, -no exact method of inference exists because it is impossible to determine the exact
degree of freedom
-two alternative approximations:
1. Welch's Approximation of Degrees of Freedom
2. Satterwaite's Approximation of Degrees of Freedom
Welch's Approximation of Degrees of Freedom
take the smaller number of observations (n1 or n2), then subtract to determine the
degrees of freedom for the t-test
-best method by hand
Satterwaite's Approximation of Degrees of Freedom
a formula usually used by the computer
**MORE ACCURATE calculation
independent samples
Sample 1: the weights of 19 newborn females
sample 2: the weights of 19 newborn males
independent samples
sample 1: the score of 13 students who took the ACT
sample 2: the scores of 13 other students who took the SAT
paired or dependent samples
sample 1: the pre-training weight of 17 athletes
sample 2: the post training weights of the same 17 athletes
ANOVA
