STATISTICS

Question 1

variable that is presumed to cause a change in another variable

Answer 1

independent variable

Question 2

categorical variable

Answer 2

RELIGION

Question 3

interval data

Answer 3

corresponds to a NUMBER

1) DISCRETE
2) CONTINUOUS
opposite of categorical data

Question 4

discrete?

Answer 4

even numbers ONLY (aka number of toes)

Question 5

continuous?

Answer 5

constant scale….lots of decimal

Question 6

categorical data types?

Answer 6

1) dichotomous
- male/female
- dead/alive
2) nominal
- no ranking; blue eyes
3) Ordinal
- ranked data, but NO consistent scale (ex - ASA class, but hard to quantify difference between them)

Question 7

cannot use mean/median for which kind of data?

Answer 7

categorical data (eye color)

Question 8

weakness for Mean?

Answer 8

outliers

symmetrical data

Question 9

median useful for what kind of data?

Answer 9

Skewed data

Question 10

can’t use mode with what sort of data?

Answer 10

continuous (too many decimal points, won’t have any repeating)

Question 11

variance?

Answer 11

average of SQUARED deviations from the mean

Question 12

easier than variance?

Answer 12

-standard deviation

Question 13

standard deviation relates to variance how?

Answer 13

square root of variance

Question 14

how much within 1 standard deviation of mean?

Answer 14

68%

Question 15

how much within TWO standard deviations of mean?

Answer 15

95%

Question 16

three SD?

Answer 16

99.7%

Question 17

Probability ASSUMES:

Answer 17

-normal distribution

Question 18

99% confidence interval has ____ range of values than 95% CI

Answer 18

LARGER (more confidence, larger range)

Question 19

null hypoth: defendant is innocent

Answer 19

alternative: defendant is guilty

Question 20

rejecting the null

Answer 20

the defendant is guilty

Question 21

TYPE 1 ERROR

Answer 21

false positive

convict when innocent
(you’re pregnant…as a male)

Question 22

how type 1 error relate to Null?

Answer 22

reject a TRUE Null

Question 23

type II error relate to null?

Answer 23

don’t reject false null

guilty but gets off

Question 24

statistical measure of the strength of the evidence?

Answer 24

P-value

Question 25

alpha is the probability of WHAT?

Answer 25

type 1 error

Question 26

if alpha is .05, what's that mean?

Answer 26

5/100 chance a given result occurred purely by chance

Question 27

ways to DECREASE type 2 error?

Answer 27

- increase alpha - increase sample size (MOST COMMON) - decrease sample variability - increase difference measured btwn the compared groups

Question 28

not noticing an error when one's there?

Answer 28

type 2 error

Question 29

P-value is the likelihood of Type __ error?

Answer 29

Type 1

Question 30

parametric data (categorical things that u can measure) , use THIS test?

Answer 30

T-test (comparing two different things): ONE-SAMPLE: -comparing different things on the same object TWO-SAMPLE (unpaired) -comparing something, but on two different objects

Question 31

ANOVA (analysis of variance) used for?

Answer 31

when many different tests performed, but would be too many samples and too much error --puts all data into one number (F) and gives you one probability for the null hypothesis

Question 32

non-parametric data is what?

Answer 32

NOT NORMAL DISTRIBUTION | -or ordinal/categorical (eye color, ASA class)

Question 33

tests for non-parametric data?

Answer 33

- Mann-whitney U test - Wilcoxon T-test - Kruskall Wallis H-test - Friedman x2 test

Question 34

Chi-square tests for?

Answer 34

TWO things are independent

Question 35

correlation analysis?

Answer 35

strength of relationship between variables (pos or neg)

Question 36

Regression analysis?

Answer 36

mathematical equation to describe relationships of variables

Question 37

R-square tells us?

Answer 37

Proportion of variability in Y accounted for by X --aka, STRENGTH of relationship btw variables x&y

Question 38

R-square values?

Answer 38

``` 1 = perfect linear relationship 0 = no linear relationship ```

Question 39

multiple regression used when?

Answer 39

describe multiple independent variables are related to SINGLE dependent variable

Question 40

ANOVA

Answer 40

multiple variables, but CATEGORICAL data

Question 41

ODDs ratio:

Answer 41

corresponds to prob of event occurring vs not occurring 1 - more likely in first group

Question 42

RISK ratio

Answer 42

calculated same way as OR

Question 43

diff btw OR and RR?

Answer 43

OR = no implication of temporality of association; RR = risk of developing one condition IF you have exposure

Question 44

OR uses what studies?

Answer 44

CASE CONTROL

Question 45

RR uses what studies?

Answer 45

COHORT studies

Question 46

PPV calc?

Answer 46

true pos test/all positive TESTS

Question 47

NPV calc?

Answer 47

true neg test/all neg TESTS

Question 48

sensitivity?

Answer 48

pos test with actual dz/total have dz

Question 49

specificity?

Answer 49

neg test and no dz/total neg dz