bias is measureable or nonmeasurable?
nonmeasurable
random error is measurable or nonmeasurable?
measurable
convenient sample is often a _______
bias sample
relationship of sample size to random error
larger sample size, smaller random error
considerations in sample
- is the sample representative of the population
- is the sample large enough
- are there other variables that explain the results
independent variable
presumed cause
dependent variable
outcome from the presumed cause
extraneous variables
must be kept constant
nominal scale
no order, only identifiers
ordinal scale
some order exists
scales of measurement
- nominal
- ordinal
- ratio
- interval
process of testing a statement about the population against an alternative statement
hypothesis testing
if p > a,
null hypothesis is not rejected
if p < a,
null hypothesis is rejected
two tailed hypothesis
null: mean is equal to X
alternative: mean is greater than or lesser than X
one tailed hypothesis
null: mean is greater than or equal to X
alternative: mean is less than X
the null and alternative hypotheses must be phrased such that they are:
mutually exclusive
collectively exhaustive
rejection of the null hypothesis means:
there is a clinically important difference
non-rejection of the null hypothesis:
there is no clinically important difference
type 1 error
concluding a difference when there is none
type 2 error
concluding the absence of a difference when there is a difference
what type of error is more serious
type 2
what type of estimate is preferable, point or confidence interval
confidence interval
