Probability and Statistical Significance Flashcards Preview

RUSVM Epidemiology 2016 > Probability and Statistical Significance > Flashcards

Flashcards in Probability and Statistical Significance Deck (40):
1

What are the two way studies can have errors?

Caused by chance (Random Error)
Not Causes by chance (Bias or Systematic error)

2

Random Error

Due to Chance
Does not bias a study

3

Systematic error

Due to Bias
Error that is inherent to the study method being used and results in a predictable and repeatable error for each observation

4

'Statistically significant'

if a study result is unlikely to have occurred by chance

5

A Chance occurrence

Something that happens unpredictably without discernible human intention or with no observable cause: Caused by chance or random variation

6

Random Variation

there is an error in every measurement. If we measure something over and over again, we will get slightly different measurements each time and a few measurements may be extreme

7

Statistical inference

if we measure something only once, how sure are we that our measurement has been caused by chance

8

What are the two methods used for estimating how much random variation there is in a study and whether the result was likely to have been caused by chance?

Confidence Intervals
P-values

9

Confidence Intervals

Estimates how much random variation there is in our measurement
The range of values where the true value of our measurement can be found (assuming the study wasn't biased)
For some measures it can estimate whether the measure was likely to have been caused by chance

10

P-values

Used to estimate whether the measure was likely to have been caused by chance or not

11

If the 95% Confidence Interval odds ratio DOES NOT INCLUDE ONE....

the odds ratio is statistically significant

12

When the odds ration is equal to 1

there is no association

13

Odds Ratio is greater than one and confidence interval doesn't include one

Positive association
Statistically significant

14

Odds Ratio is greater than one, but confidence interval includes one

No association
Not statistically significant

15

Odds Ratio is less than one and confidence interval doesn't include one

Negative association
Statistically significant

16

Odds Ratio is less than one and confidence interval includes one

No association
Not statistically significant

17

What value must the P-value be less than to be statistically significant?

0.05

18

Null Hypothesis

The 'hypothesis of no association' - this is no association between exposure and disease

19

Alternative hypothesis

The 'research question' - that there is an association between exposure and disease

20

If the p-value is less than 0.05

Reject the null hypothesis and accept the alternative hypothesis

21

If the p-value is greater than 0.05

Accept the null hypothesis and reject the alternative hypothesis

22

Type I (alpha) error: False Positive

Rejecting the null when it is NOT false
p=0.05

23

Type II (Beta) error: False Negative

Not rejecting the null when it is false (an association truthfully exists)
Conventionally set at 0.20

24

Categorical

The variable is broken into discrete categories

25

Nominal

Named (not ordered)

26

Ordinal

Named and ordered but no constant value between ranks

27

Continuous

The variable is numeric and can have any one of many possible values

28

Mean

Average
Sum of all values/ # of data points
VERY sensitive to extreme values

29

Median

The value which is the center, with half of the data points "below" and half of the data points "above"

30

Mode

The most frequently occurring value/observation

31

Dispersion

Describes how closely the values are gathered around the center of distribution

32

What are the two measures of dispersion?

Range
Standard Deviation

33

Range

the difference between the minimum and maximum

34

Standard Deviation

The average distance between each measurement and the mean

35

What are the two statistical tests?

Chi-squared test
Student's T-Test

36

Chi-squared test

difference in proportions
Used for categorical data
All two by two tables

37

Student's T-test

difference in means
Compares the average of two groups
used for continuous data

38

Correlation

Measures the strength ad direction of a linear relationship between two continuous variables

39

Is an r value greater than 0.80 strong or weak correlation?

Strong

40

Is an r value less than 0.80 strong or weak correlation?

Weak