module 5 hypotheses Flashcards
point estimation
type of parameter estimation
- one value: mean, mode
- margin of error
interval estimation
type of parameter estimation
- more than one point, consists of a range of values within which the population parameter is thought to be
null hypothesis
proposes that there is no difference or relationship between the variables of interest.
“there will be no difference”
directional hypothesis
states that there will be a relationship between two variables and gives the expected direction of that relationship.
“ will have more or less”
non-directional hypothesis
states that there will be a significant relationship between two variables, but the direction is not stated.
p value (probability value)
represents the probability that the results were obtained by chance alone
alpha level
benchmark value set by researcher before testing
- .05, .10, .01
- will happen 5 times out of 100
reject the null
researchers believe that the variables are significantly associated with each other.
p<a></a>
Accepting the null
researchers believe that the variables are not significantly associated.
p>a
Type I error
false positive
- you think there is a difference but there is not.
- null hypothesis rejected when it is true
- likelihood defined by alpha
Type II error
False negative
- Accept a null that is false
- you think there is no difference, when there really is
- likelihood of making this error is beta (20%) reasonable
effect size
magnitude of the relationship between variables: how much the intervention made an effect.
- Small, medium, large effect sizes
- small effect size requires larger sample size
power
ability to detect statistically significant differences
- 1 - Beta (20%)
- power analysis can be done before or after study; determine sample size.
necessary quantities for power analysis
alpha: .05
power: 80%
effect size: small, medium, or large
sample size
- if you have 3 you can determine the 4th
tight controls and balance
tight control -> less Type 1 risk, inc. type II risk
