Stats Flashcards

1
Q

Mean

A

Average add sum/number

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Median

When is it preferred?

A

Rank lowest to highest

Middle value

Preff: in ordinal and continuous data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Mode?

A

Number than happens most frequently

Pref for nominal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Range?

A

Difference between largest and smallest

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Normal distribution note

A

Mean median and mode are the same

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Where do values fall in normal distribution?

A

68% within 1 SD

95% fall within 2 SDs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the possible error margin called?

A

APLHA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Alpha commonly set?

A

5%

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

When is the null rejected?

A

When the P-value is <0.05 if that is where the alpha is set

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

CI equation>

A

CI=1-a

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

If a CI contains what?

A

0 not statistically significant

FOR MEANS

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Ration data? Three examples?

A

OD, HR, RR

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

CI confidence interval in ration data?

A

If it includes 1 not statistically significant

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Type 1 errors? Also called?

A

Falso positive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Type 1 error in relation the H0 and Ha?

A

Alternative hypothesis was accepted and Null rejected in error

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Type II is also called? How is it denoted?

A

False negative

B

Alternative was rejected and Null accepted in error

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is study power?

A

Probability that the test will reject the null correctly

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

How is power calculated?

19
Q

RR values and interpretation? 3

A

=1 no difference

>1 greater risk of outcome in treatmetn

<1 lower risk in Tx

20
Q

NNT NNH rounding

A

NNT anything greater than a whole number is rounded up to the next whole

NNH: anything greater than a whole number is rounded down

21
Q

When is ANOVA used?

A

When using continuous data with 3 or more variables

22
Q

T-Test

A

continuous data that is normally distributed

Student t-test: control versus placebo

23
Q

Continuos data 2 tests

A

T-test and ANOVA

24
Q

Discrete or categorical data test used? 1

A

Chisquared

25
Types of continuous data?
Ratio and Interval
26
Types of discrete or categorical data? 2
Nominal: arbitraty raking Ordinal: ranked in logial order
27
Types of correlation tests?
Spearmans ranked order correlation Pearsons correlation coefficient
28
What is the primary correlation test for continous data?
Pearson correlation coeffiecient
29
A test with 100% sensitivity?
Test will be positive in all patients who have the condition
30
Specifity
100% will be negative in all pts without the condition
31
What do forest plots do?
provife CIs for differnce data or ratio data
32
Forest plot interpretation
Vertical line is called the line of no effect set to zero for difference data and 1 for ration data Hortizontal line is the length of the confidence interval
33
Case control study What does it do? Waht data?
Compared pts with a dx to those without Retrospective * ORs
34
Cohort study? What data?
Compares outcomes of 1 group of pts exposed and not exposed Prospective or retro but less common 1. OR
35
Cross sectional study? What data?
Relationship at a point in time variable and outcome at one point in time a CROSS SECTION 1. p values
36
Cae report and case series? What data?
Unique thing that happens in one patient or a few patietns No statistical validity
37
RCTs What data? What data?
Compares control adn placebo HR CI
38
Parallel RCT?
Randomized to tx and control for the entire study
39
Cross over RCT?
Randomised to one or two tx sequences P values
40
Factorial design?
Randomized to more than the usual two groups
41
Meta analysis
Results from multiple study greatest signficantce power
42
sYSTEmmic review article?
Summary of lit that focuses on a specific question or topic
43