chapter 10 Flashcards
the 2 samples of n1 form pop 1 and n2 from pop 2- the 2 samples taken separately and independently are called
independent simple random samples
What is the point estimator of the difference b/w 2 pop means
xbar 1 - xbar 2
what is the random sample size adequate for most applciaitons of interval estimation and Hype test procedures
n1 greater than or = 30
n2 greater than or = 30
when one or both sample sizes are less than 30 for the difference b/w 2 pop means
the distribution of the pop becomes important
when using the t-distribuiton for typ test Q1 and Q2 unknown
remember to always round the df down before using the t-distribution
whenever possible what sizes of n1 and n2 are recommended when Q1 and Q2 are not known
equal sizes of n1 and n2 are recommended
- the total samples size is at least 20 (n1 + n2)
- can be expected to provide very good results even if the pops are not normally distributed
what is the other approach when Q1 and Q2 are not known
- the two sample sd are combined to provide the pooled sample variance
how do you calculate the degrees of freedom for a pooled variance
n1 + n2 -2
what is the problem with the pooled sample variance
it is often difficult to verify that both sd are equal
- the pooled sample variance may not provide satisfactory results especially if the sample size are quite different
what is the problem with the pooled sample variance
it is often difficult to verify that both sd are equal
- the pooled sample variance may not provide satisfactory results especially if the sample size are quite different
describe the independent sample design
a simple random sample of workers is selected and each worker in the sample uses method 1
- a second independent simple random sample of workers is selected using method 2
- test of the diff b/w 2 pop means is based on procedures used in prevoiusly
describe the matched sample design
- one SRS of workers is selected
- each worker 1st uses one method and then uses the other method
- the order of the 2 methods performed is assigned randomly
- each worker provides a pair of data values
- each method is tested under similar conditions
- this often leads to a smaller sampling error than independent sample design b/c same workers are sued for both methods
which method (independent or matched design) leads to smaller sampling error
matched sample
what is the key to the matched sample design method
we only consider the column of differences
what is the formula for test stat for hyp test involving matched samples
t = dhat - Md / Sd/ square root n