week 4 Flashcards Preview

EBP > week 4 > Flashcards

Flashcards in week 4 Deck (19)
Loading flashcards...
1

bias occurs if certain participants have a/an ________ probability of being selected

unequal

2

random sampling examples

simple random
stratefied
cluster

3

non-probability sampling examples

convenience sampling
quota sampling
snowball sampling

4

standardized normal distributeion

gaussian distribution

5

z score and its equation

labeled by number of standard devs

Z=(X-(MEAN OF X))/ standard dev

6

percentile equation

(# of scores less than x)/N x 100

7

inference

making generalizations about a pop based on data from samples

use hypotheses to make inferences

8

null hypothesis

no association
no difference in means
no difference in proportions

9

alternate hypothesis

there is an association
diff in means or proportions

10

alpha

the probably that the investigator accepts a priori that the results may be due to chance rather than a true difference

accepted proability of a type 1 error

11

p value

the actual probability from the data that results are due to chance rather than a true difference

the probability that a null hypothesis is true

12

reject the null if

p

13

type I error

recommend/ provide treatment that is inneffective

14

type II error

failure to identify an effective treatment that may benefit your patient

15

confidence interval

calculated range of values around the estimated measure that has a specified (95%) probability of including the true value of the measure in the population

may be used to determine significant, or precision

based on standard error

16

determining if means are different using CIs

is the difference between the two means falls between the CI, they are not different

17

how precise the estimate is depends on the ____ of the CI

width

18

power

probability of a type 2 error
related to variability of measure in question
magnitude of the efect
larger sample size-> greater power

19

post hoc analysis

useful if findings were null or the sample size was small
enables the researches to estimate the probability of a type II error