ch 12

Question 1

random error?

Answer 1

errors that rise from chance that lead to an incorrect estimate of association

Question 2

chance

Answer 2

uncontrollable force with no apparent cause that arises du to an unforeseeable and unpredictable process

Question 3

sources of random error

Answer 3

• This can occur in all types of epidemiological studies
• Caused by measurement error, sampling variability
• All studies have some random error

Question 4

effects of random error

Answer 4

• Creates the appearance of an association when there is none or masks an association that really exists
• This can be described in the analysis

Question 5

measurement error

Answer 5

• Non-systematic measurement errors can lead to random error
• Recall that measurement errors occur when there are mistakes in ascertaining exposures or outcomes
• Measurements may vary due to subtle differences in procedures that are difficult to control

Question 6

sampling variability

Answer 6

• Study subjects are a sample from a population of interest
• Sampling implies variability. ‘
• Random occurs when selected subjects do not represent the underlying population just by chance.

Question 7

how can we reduce random error to increase precession ?

Answer 7

• Use an accurate measurement tool, repeat measurements
• Increase sample size
• Use an efficiency study design that maximizes the mount of information obtained for a given sample size

Question 8

precision

Answer 8

lack of random error, state of being precise or exact

Question 9

two ways we can calculate random error

Answer 9

1. Hypothesis testing (p-values)
2. Confidence interval

Question 10

three steps in Hypothesis testing?

Answer 10

1. . Specify a null and alternative hypothesis
2. Use a statistical test to determine the compatibility of study results with the null hypothesis (t-test, chi-square test, others)
3. . Decide whether or not to reject the null based on the P-value association