Statistics

Question 1

Type I error

Answer 1

rejects null hypothesis incorrectly i.e. falsely assumed there was a difference when no difference exists

Question 2

Type II error

Answer 2

accepts null hypothesis incorrectly (usually from small sample size) i.e. the outcomes are interpreted as equal when there is actually a difference

Question 3

Null hypothesis

Answer 3

hypothesis that no difference exists between groups p < 0.05 rejects the null hypothesis

Question 4

p < 0.05

Answer 4

equals > 95 % likelihood that the difference between the populations is true, < 5 % that the difference occurred by chance alone and is not true

Question 5

Variance

Answer 5

spread of data around a mean

Question 6

paramenter

Answer 6

populations

Question 7

mean, median, mode

Answer 7

average, middle value, most frequently occuring

Question 8

randomized controlled trial

Answer 8

prospective study with random assignment to treatment and non-treatment groups. AVOIDS treatment biases

Question 9

double - blind controlled trial

Answer 9

prospective study in which patient and doctor are blind to the treatment. AVOIDS observational biases

Question 10

Cohort study

Answer 10

prospective study, compares disease rate between exposed and unexposed groups

Question 11

case controlled study

Answer 11

retrospective study in which those who have the disease are compared with a similar population who do not have the disease, the frequency of the risk factor is then compared between the two groups

Question 12

meta analysis

Answer 12

combining data from different studies

Question 13

t test

Answer 13

2 independent groups and variable is quantitative –> compares means (mean weight between 2 groups)

Question 14

paired t test

Answer 14

variable is quantitiatve, before and after studies (weight before and after, drug vs placebo)

Question 15

ANOVA

Answer 15

comapres quantitative variables (menas) for more that 2 groups

Question 16

nonparametric statisitcs

Answer 16

compare categorical (qualitative) variables (race, sex, medical problems and disease, medications)

Question 17

chi-square test

Answer 17

compares 2 groups with categorical (qualitative) variable (number of obese patients with and without diabetes versus number of non-obese patients with and without diabetes)

Question 18

kaplan- meyer

Answer 18

small groups –> estimates survival

Question 19

relative risk

Answer 19

= incidence in exposed/ incidence in unexposed

Question 20

power of test

Answer 20

= probability of making the correct conclusion = 1- probability of type II error
larger sample size increase the power of a test
likelihood that a test is true

Question 21

prevalence

Answer 21

number of people with disease in a population (e.g. number of patients in US with colon CA)
long standing diseases increase the prevalence

Question 22

incidence

Answer 22

number of new cases diagnosed over a certain time frame in a population

Question 23

sensitivity

Answer 23

ability to detect a disease (SNOUT) (rules dz out)
high sensitivity - a negative test result means a patient is unlikely to have disease
TP/ TP + FN

Question 24

specificity

Answer 24

ability to state no disease is present (SPIN) (rules dz IN)
High specificity - a positive result means pt likely to have disease
TN/ TN+ FP

Question 25

Positive predictive value

Answer 25

likelihood that a positive result PT actually has the disease
true positives/ TP + FP

Question 26

Negative predictive value

Answer 26

likelihood that a negative result PT does NOT have the diesease
true negatives/ TN + FN

Question 27

accuracy

Answer 27

TP + TN/ TP+ TN + FP+ FN