Week 5 VanNess- Introductory Stats

Question 1

what is the null hypothesis

Answer 1

prediction that the observed difference is due to chance alone and not due to a systematic cause

Question 2

what is the alternate hypothesis

Answer 2

prediction that some observed difference is significant and due to some knowable cause

Question 3

what is a type 1 error

Answer 3

false positive
-alpha

Question 4

what is a type 2 error

Answer 4

false negative
-beta

Question 5

what does statistical testing test

Answer 5

likelihood of differences occuring by chance alone

Question 6

what is the p-level

Answer 6

the predetermined probability researcher is willing to make a type 1 error
- indicates difference between 2 groups

Question 7

what is the p-level also referred to as

Answer 7

the alpha level

Question 8

what does p<0.05 mean

Answer 8

only 5% of the time, the difference will be observed due to chance alone
-95% confident that results were due to independent variables

Question 9

what are the levels of measurement

Answer 9

-nominal
-ordinal
- interval
-ratio

Question 10

what is nominal and give examples

Answer 10

label or category without rank
-ex: gender, hair color, living accommodation, religion

Question 11

what is ordinal and give examples

Answer 11

label or category with some meaningful order or sequence
-ex: class ranking, socioeconomic status, likert scale (agree, disagree), level of agreement (yes or no)

Question 12

what is interval and give example

Answer 12

scaled measure with an arbitrary zero point
- ex: temperature (F or C), IQ, SAT scores

Question 13

what is ratio and give example

Answer 13

scaled measure with an absolute zero point
-ex: weight/height
-income earned in a week
- years of education
-number of children

Question 14

what tests would you use for nominal date values in 2 independent groups, 2 dependent groups, 3+ independent groups, and 3+ dependent groups?

Answer 14

-2 independent groups - chi square
- 2 dependent groups- McNemar
- 3+ independent groups- Chi Square
- 3+ dependent groups - cohcran’s Q

Question 15

what tests would you use for ordinal data values in 2 independent groups, 2 dependent groups, 3+ independent groups, and 3+ dependent groups?

Answer 15

-2 independent groups - Mann- Whitney
- 2 dependent groups- Wilcoxon Matched Pairs
- 3+ independent groups- Kruskal-Wallis
- 3+ dependent groups - Friedman’s Chi square

Question 16

what tests would you use for interval/ratio data values in 2 independent groups, 2 dependent groups, 3+ independent groups, and 3+ dependent groups?

Answer 16

-2 independent groups - independent t-test
- 2 dependent groups- dependent t-test
- 3+ independent groups- ANOVA
- 3+ dependent groups - repeated measures of ANOVA

Question 17

what do inferential stats not do

Answer 17

-prove cause and effect
-estimate clinical effectiveness
- estimate risk/benefit

Question 18

what do inferential stats do

Answer 18

• estimate the probability of getting results due to chance
• suggest numerical differences

Question 19

what does effect size describe

Answer 19

the magnitude of difference

Question 20

what do you need to calculate effect size

Answer 20

mean and standard deviation for each group

Question 21

what does standardized mean effect express

Answer 21

the mean difference between the 2 groups in standard deviation units (Cohen’s D)

Question 22

what is the range of values for effect size

Answer 22

-3.0 to 3.0 like SD

Question 23

what are the ranges for large, moderate and small effect size values

Answer 23

large- 0.8
moderate- 0.5
small- 0.2

Question 24

what is relative risk

Answer 24

ratio of probability of the event occurring in the exposed group vs non-exposed group

Question 25

what is the formula to measure relative risk when the exposure is negative (toxin)

Answer 25

incidence rate of exposed/incidence rate of unexposed

Question 26

what is the formula to measure relative risk when the exposure is positive (prenatal care)

Answer 26

incidence rate of unexposed/ incidence rate of exposed

Question 27

what is the odds ratio

Answer 27

estimate of association

Question 28

what does the odds ratio compare

Answer 28

the odds of an event in one group to the odds of an event in some comparison group

Question 29

how are relative risk and odds ratio similar

Answer 29

both the odds ration and relative risk compare the likelihood of an event occurring between two distinct groups

Question 30

what is the advantage to relative risk over odds ratio

Answer 30

-RR is easier to interpret and consistent with the general intuition
-compares between subgroup and population rather than subgroup and the remainder of the population

Question 31

what limits RR calculation

Answer 31

case-control designs

Question 32

when are OR and RR comparable

Answer 32

when the disease being studied is rare

Question 33

what happens with OR and RR when the disease is more common

Answer 33

OR overestimates

Question 34

what is sensitivity

Answer 34

the proportion of patients with the disease who test positive

Question 35

what is specificity

Answer 35

the proportion of patients without the disease who test negative

Question 36

what does the confidence interval tell you

Answer 36

the most likely range of the unknown population average

Question 37

what impacts the width of the confidence interval

Answer 37

-confidence level usually 95%
- variability: standard deviation
- sample size: smaller sample sizes generate wider intervals

Question 38

how do you calculate odds ratio

Answer 38

odds of people with disease being exposed

Question 39

what would 2 standard deviations above and below the mean indicate

Answer 39

a result attributable to something other than chance alone