Term
|
Definition
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
|
Definition
Mean, median & mode are equal.
68% of value falls under first standard deviation, 95% fall under second standard deviation, 99% fall under third standard deviation |
|
|
Term
Measures of Dispersion:
Range |
|
Definition
Measures of Dispersion:
Range - Difference between highest and lowest observation |
|
|
Term
Measures of Dispersion:
Standard deviation |
|
Definition
Measures of Dispersion:
Standard deviation - A measure of each score from the mean |
|
|
Term
Measures of Dispersion:
Variance |
|
Definition
Measures of Dispersion:
Variance - Square of standard deviation |
|
|
Term
| Standard error of the Mean (SEM) |
|
Definition
Used for confidence limit determination.
As sample size increases, SEM decreases |
|
|
Term
Measures of Shape:
Positive Skew |
|
Definition
Positive Skew - Tail on the right hand side
Mean > Median |
|
|
Term
Measures of Shape:
Negative Skew - |
|
Definition
Measures of Shape:
Negative Skew - Tail on the left side
Mean < Median |
|
|
Term
Statistical Significance -
Null hypothesis |
|
Definition
Statistical Significance -
Null hypothesis: States that there is no difference between things being compared in regard to a specific result |
|
|
Term
|
Definition
| Alternate Hypothesis: There is a difference between the items being compared, with regard to a specific result |
|
|
Term
|
Definition
| Probability that the results are not due to chance. Eg. 95% CI - means 5% chance results are due to luck |
|
|
Term
|
Definition
Types I (alpha error)
False Positive: Difference is calculated when not really present |
|
|
Term
|
Definition
Type II (Beta error):
False negative:
Fail to reject null hypothesis, when in fact it is true |
|
|
Term
|
Definition
| How often outcome of interest occurs in control Group |
|
|
Term
| Experimental Event Rate (EER) |
|
Definition
| How often outcome of interest occurs in Experimental population group |
|
|
Term
|
Definition
Amount of risk removed by the intervention compared to the control.
(EER/CER) |
|
|
Term
| Absolute Risk Reduction (ARR) |
|
Definition
Absolute Risk Reduction (ARR):
Number of people spared by using intervention.
CER-ERR |
|
|
Term
| Relative Risk Reduction (RRR) |
|
Definition
Percentage of baseline link removed
RRR = ARR/CER |
|
|
Term
| Number Needed to Treat (NNT) |
|
Definition
Number of subjected needed to be treated to prevent one event.
1/ARR |
|
|
Term
|
Definition
(a/b)/(c/d)
Ratio comparing percentage of treatment group outcomes to control group outcomes.
of OR > 1, event occurs more in the treatment group. OR < 1, event occurs more in the control group. |
|
|
