Term
|
Definition
- the probability that the statistical results are correct; based on the normal distribution and the “68-95-99.7 Rule” |
|
|
Term
|
Definition
- a range of numbers within which we expect the true prevalence to lie; based on level of confidence (usually 95%) |
|
|
Term
|
Definition
- A percentage representing our confidence that our result is not by chance; expressed as a percent (often 95% is used) |
|
|
Term
|
Definition
- rejecting a null when the null is true (based on statistical results); (the researcher rejects the null and concludes that a statistically significant difference exits, HOWEVER, no true difference is present) |
|
|
Term
|
Definition
- : failure to reject the a null when the null is true; (the researcher concludes that no statistically significant difference exists and “accepts” the null HOWEVER, a significant difference DOES exits) |
|
|
Term
|
Definition
- An expression of chance that an event will occur. Ranges from 0 (no chance) to 1 (certainty) |
|
|
Term
|
Definition
- A numerical or mathematical expression; a limit for rejecting the null hypothesis or failing to reject the null; assesses whether the difference between observations is a coincidence or a true difference; usually stated as 0.01, 0.05, 0.01, or 0.0001. |
|
|
Term
Statistically Significant |
|
Definition
- probability less than 0.05 (p<.05) - null rejected = there is a difference |
|
|
Term
not statistically significant |
|
Definition
- probability greater than 0.05 (p>.05) - failed to reject null = there is no difference |
|
|
Term
non parametric statistical tests |
|
Definition
- used when differences are looked at in categories (not “how much are they different” but “are they different”); data comes from a population that is not normally distributed and is nominal or ordinal; sample is small |
|
|
Term
parametric statistical tests |
|
Definition
- used on interval/ratio data from a normally distributed population; looks at differences numerically (quantitatively); sample is large and random |
|
|