Errors Flashcards
1
Q
Random errors
A
- fluctuations in either direction in the measured data due to the precision limitations of the measurement device
- often due to the researcher’s inability to take a measurement in the same way and get exactly the same result
2
Q
Systematic errors
A
- also called bias
- these are reproducible errors that are consistently in the same direction
- a problem which persists throughout the entire experiment is called bias
3
Q
Type 1 error
A
- incorrect rejection of the null hypothesis (false positive claim in favour of the research hypothesis
- likelihood of type 1 error is called alpha
- as long as it is less than 0.05 its ok and is less than the level of statistical significance
4
Q
What can lead to type 1 error?
A
-repeated testing of hypothesis using the same data, multiple subset analysis and secondary analysis as at least one test will be positive out of 20 if p is 0.05
5
Q
Type 2 error
A
- the incorrect acceptance of the null hypothesis, concluding no differences exist between two treatments actually exist (flase negative rejection of research hypothesis)
- probability of this is called beta
- usually due to small sample size or large variance
- 1-beta is the power of a study and is the ability to obtain a true positive
6
Q
Power
A
- ability to obtain a true-positive result
- power= 1- B
7
Q
Probability of type 1 error
A
-alpha or p value
8
Q
Probability of type 2 error
A
-beta
9
Q
Alpha
A
5%
10
Q
Beta
A
20%
11
Q
Power
A
80%
