hypothesis testing Flashcards

1
Q

how do you do a binominal test for hypothesis testing?

A
  • binominal test is a statistical test that concerns whether a proportion observed in your data is different from a known proportion
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

what can a statical test systematically test?

A
  • tests whether a given scientific claim is valid or not
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

what are answers like in most cases in scientific research?

A
  • no definitive answers to any kind of questions so scientists prefer to discuss based on probability
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

when do we reject the hypothesis?

A
  • if the probability is very low
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

when do we accept the hypothesis?

A
  • if the probability is not so low
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

what are p -values?

A
  • probabilities used to reject hypotheses
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

what are the threshold levels for p- values? when are they defined?

A
  • alpha level
  • normally 0.05
  • needs to be defined prior to a study
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

what p - value do you reject the hypothesis at?

A
  • if p- value < 0.05
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

what p -value do you accept your hypothesis at?

A
  • if p- value > 0.05
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

what is a null hypothesis?

A
  • hypothesis against the research question, claiming that there is no difference in the result
  • only differences observed are just noise/ error
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

what is the symbol for null hypothesis?

A

H 0

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

what do scientifistic often have to struggle against?

A
  • struggle against the probability that the observed difference could be chance due to noises
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

what is the research/ alternative hypothesis?

A
  • claims that there is a difference in the result
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

what is the symbol for research hypothesis?

A

H a

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

in statistics what do we test?

A
  • test the probability that the null hypothesis is true
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

why don’t we test the research probability?

A
  • because you can never prove something is true
  • can prove the null hypothesis is false
17
Q

what do we do in hypothesis testing?

A
  • first define H0
  • make an attempt to reject this by showing there is a statistically significant difference
18
Q

how is significance estimated?

A
  • by the probability that the difference occurred by chance (p- value)
19
Q

what always occurs when dealing with probability?

A
  • errors
  • two types; 1 and 2
20
Q

what is type 1 error?

A
  • false positive
  • reject the null hypothesis when it is true
21
Q

what is type 2 error?

A
  • false negative
  • not to reject the null hypothesis when it is false
22
Q

do we want to minimise one of the types of errors more?

A
  • researchers need to minimise both but based on your research question you may want to focus on one type more than the other
23
Q

what are the three typical types of tests?

A
  • observed proportion< expected proportion
  • observed proportion > expected proportion
  • observed proportion = expected proportion
24
Q

described observed proportion> expected proportion

A
  • 1 - cumulative probability from observed to max
24
Q

describe observed proportion < expected proportion

A
  • cumulative probability from 0 to observed
25
Q

describe observed proportion = expected proportion

A
  • two tailed cumulative probability same distance from the mean
26
Q

how do you conduct a binominal test in hypothesis testing?

A
  • describe null hypothesis with expected proportion
  • report observed proportion from data
  • report probability (p- value) that the null hypothesis is true
  • optionally report the confidence interval
27
Q

what is the confidence interval?

A
  • range of plausible values associated with a confidence level
28
Q

what is the normal confidence interval and what does this mean?

A
  • normally 95%
  • means you are 95% certain that true proportion falls within CI
29
Q

what does confidence interval include?

A
  • often used to the range in which true proportion may lie