module 7 Flashcards
null hypothesis
- skeptical viewpoint of the research question
- devils advocate
Null hypothesis is typically written as:
a) Ha: statement
b) Ho: statement
c) Ho: numerical values
d) Ha: categorical values
b)
alternative hypothesis
- positive viewpoint of research question
- pattern you want to exist
- contains everything not covered by null hypothesis
Alternative hypothesis is typically written as:
a) Ha: statement
b) Ho: statement
c) Ho: numerical values
d) Ha: categorical values
a)
How should you write your null and alt hypothesis after you have written it out in words
- convert the hypothesis to symbols for stat clarity
- ex. Ho: X1=X2 and Ha: X1≠X2
t or f: the null and alternative are mutually exclusive.
true
t or f: the alternative hypothesis describe all possible outcomes alone
false, The null hypothesis and alternative hypothesis describe all possible outcomes
The ______ hypothesis always includes the equality statement
null
directionality
- if there is direction in the null and alt hypotheses in terms of the measurement variable
________ hypotheses just state that there should be a difference in the alternative hypothesis, whereas ________ hypotheses state that the difference should be in a specific direction
nondirectional, directional
null distribution
- sampling distribution that you get if you repeatedly sample from fictitious statistical population where null hypothesis is true
- The null distribution is the sampling distribution of an imaginary statistical population where the null hypothesis is true
The first step in a hypothesis test is to define the ________, and the second step is to establish the ________.
define the null and alternative hypothesis, establish the null distribution
what are the two possible outcomes of a statistical test
- it is likely your data could come from the null distribution, you fail to reject the null hypothesis
- it is unlikely that your data could come from the null distribution, you reject the null hypothesis
t or f: you can accept the null hypothesis as long as you have enough data to support it
false, you can only fail to reject it
what are the two probabilities used in doing a formal hypothesis test
- Type I error rate aka alpha (⍺): the probability of rejecting the null hypothesis when its true
-P-value: probability of seeing your data under the null hypothesis, area under the curve from the data to more extreme values
what is the common set point for type I error rate/alpha?
0.05/5%
who/what is the p value determined by
the data
If the p-value is less that the Type I error rate (p<⍺)…
a) you fail to reject the null hypothesis
b) you reject the null hypothesis
b)
If the p-value is greater than or equal to the Type I error rate (p≥⍺)….
a) you fail to reject the null hypothesis
b) you reject the null hypothesis
a)
wording of scientific statements should reflect what two considerations
- strength of inference
- effect size
In stating the strength of inference, essentially, you want to acknowledge that your inference is only as good as _______.
a) your sample
b) your study design
c) the data
d) all of the above
c)
t or f: effect sizes should be considered for both rejecting and failing to reject the null hypothesis
false, it is when you reject the null hypothesis
what does effect size refer to?
a) if the observed difference is meaningful for the research question
b) if the sample can be increased and have the same result
c) how big your sample was
d) if the theorized difference is meaningful for the research question
a)
t or f: Type I error is the probability of rejecting the alternative hypothesis when it is correct
false, the null hypothesis
t or f: hypothesis testing is used for the null hypothesis not the alternative hypothesis
true
what are the two main types of error rates in statistics?
Type I and Type II
Type II error rate
- failing to reject the null hypothesis when it actually is false
- probability under the alt distribution
Type I error rate
- rejecting the null hypothesis when its actually true
- probability under the null distribution
t or f: because the alternative distribution is typically unknown, the type II error is also typically unknown
true
which of the following best describes the relationship between type I and type II error rates?
a) they are independent of one another
b) as the type I error rate decreases, the type II error rate increases and vice versa
c) as the type I error rate decreases or increases so does the type II error rate
b)