2 T Test
Home
> Preview
The flashcards below were created by user
Trekofstarsx
on
FreezingBlue Flashcards
.
Quiz
iOS
Android
Members
More
By selecting a smaller alpha level, a researcher is
The z score boundaries for a critical region are determined by
the alpha level
Two possible ways of comparing two samples means
-The two means could come from two completely separate groups of participants
-Two sets of data could be gathered from the same group
Independent Measures t-test
-Two separate groups being tested
-Also known as between subjects design
t formula
[(M1-M2)-(u1-u2)]/S(M1-M2)
s
_{(M}
_{1-M2)}
_{=}
sqrt[(s1
^{2}
/n1)+(s2
^{2/}
n2)]
Estimated standard error is only true if
both ns are the same
Pooled variance formula
Pooled variance formula with standard error
see card
H0 and H1 for two independent sample means
u1-u2=0
We use the sample mean difference to
Estimate the population mean difference
Condfidence interval for two independent sample means formula
u1-u2=M1-M2+-t*S
_{M1-M2}
Two-tailed hypothesis
H
_{o}
H
_{1}
u
_{1-}
u
_{2=}
0
u
_{1}
-u
_{2}
does not equal 0
One-tailed (uppertail critical)
HO
H1
see formula sheet
One-tailed (lowertail critical)
see formula sheet
3 Assumptions of t tests
-Observations within each sample must be independent
-The population from which the samples are selected must be approximately normal
-The two populations from which samples are selected must have the same variance
Hartley F-max test
-Tests?
1.2.3.
-Tests whether our sample variances are reasonably close
1. Compute sample variance for each sample
2. Select largest and smallest of these variances and compute s2(largest)/s2(smallest) to get F max
3.Compare F-max value with the critical value
Card Set Information
Author:
Trekofstarsx
ID:
298419
Filename:
2 T Test
Updated:
2015-03-15 04:12:57
Tags:
stat
Folders:
stat
Description:
2 sample t
Show Answers:
Learn more
Home
>
Flashcards
> Print Preview