lecture 4: statistical inference - comparing two groups Flashcards
true/false: the variance of the difference (or the sum) of the 2 individual random variables equals the sum or the variances of the two populations from which the samples were drawn
true
assuming sample sizes are equal n1=n2 the variances can be averaged for greater precision to create a _________ variance
pooled
what is the formula of pooled variance
(variance of sample one + variance of sample 2)/2
for 2 samples of ______ (same/different) sizes n1 does not = n2, the pooled variance is calculated as a weighted average
different
what are the assumptions needed for a two sample t test
- random sampling
- indépendant samples
- sampling from normally distributed populations
- homogeneity of variance
does this describe a one tailed or two tailed test?
looks for an “increase” or “decrease” in the parameter
one tailed
does this describe a one tailed or two tailed test?
looks for a “change” (could be increase or decrease) in the parameter.
two tailed
this type of error occurs when the null hypothesis is rejected, when in fact, it is true; the probability of committing this type of error is usually set at 5%
type I (alpha error)
this type of error occurs when the null hypothesis is not rejected, when in fact it is false; the probability of committing this type of error is set maximally at 20%
type II (beta) error
which is worse, a type I or type II error?
type I because your saying something that doesn’t work, works
we can test for ________ by examining the ratio of the sample variance and if the quotient is less than 0.5 or greater than 2 then we need to perform a separate variance t-test
homogeneity