BIOSTATS Flashcards

1
Q

What does Bayesian statistics allow you to do?

A

Calculate conditional probabilities

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

What does the Chi squared test allow you to do? What type of answers does it use?

A

Compare results from 2 independent populations…use binary answers

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

If you have a highly sensitive test…what result are you certain is correct?

A

A negative. Like airport security…want to have highly sensitive test. You know that the people who get thru are fine. You may catch a few innocent people in the process, however.

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

If you have a highly specific test…what result are you certain is correct?

A

A positive. this would be bad for airport security. The only thing you know is that if you get a positive & catch somebody they ARE guilty. But plenty of guilty people may have sneaked thru.

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

What is the difference b/w disease prevalence & incidence?

A

prevalence–how widespread the disease is

incidence–rate of occurrence of new cases of the disease

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

What is selection bias?

A

the reality that a certain type of people are willing to participate in studies…may not be totally representative of the population

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

What are 2 ways to randomize?

A

Blocked randomization

Stratification

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

As a sample size increases, _____ increases but you also establish a more firm ________.

A

range increases when sample size increases.

You also get a steady state w/ more normal values.

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

What is internal validity?

A

How confident you are that your dependent variables in your study solely changed b/c of your independent variable.
Basically….that you had a sound study design.

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

What is external validity?

A

Extent to which the results can be generalized to other people or settings.

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

What are some things that can endanger internal validity?

A
Maturation
Testing
Instrumentation
Statistical Regression
Selection Bias
Experimental Mortality
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are some things that can endanger external validity?

A

Pretesting
Interaction
Multiple Treatments
Setting

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

What is the relationship b/w standard deviation & variance?

A

Standard deviation is the square root of variance.

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

What determines your degrees of freedom?

A

n-1

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

A peak on a graph would represent what measurement of the data?

A

the mode!!

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

If your mean is greater than your median & you have a long right tail…what is your skew?

A

Right/positive skew

17
Q

If your median is greater than your mean & you have a long left tail…what is your skew?

A

Left/negative skew

18
Q

T/F the chi square test has a larger p value than the z test.

A

False. They have the same p value.

19
Q

What question is asked with:

the predictive value of a positive result

A

What is the probability that the patient actually has this disease? They tested positive.
True Positive/True Positive + false positive

20
Q

What question is asked with:

the predictive value of a negative result

A

They tested negative. What is the probability that the patient really doesn’t have the disease?
True negatives/True negatives + False negatives

21
Q

Single subject experiments have high _____ validity, but low _____ validity.

A

High internal validity

low external validity

22
Q

What is a confidence interval?

A

CL = 1 - alpha

how confident you are that the interval contains the true average value

23
Q

What is the role of the null hypothesis?

A

it says that there is no relationship b/w 2 things…it is your job in your experiment to reject the null hypothesis.

24
Q

What is the p value?

A

ranges from 0-1
Tells us something about our null hypothesis.
If p value is low; reject the null. Yay!
If p value is high; accept or fail to reject the null. : (

25
Q

What is the t statistic?

A

Used when sample is small & it follows a normal distribution

26
Q

What is the z score?

A

compares a sample to a population…used when sample is large

27
Q

What is the alpha level?

A

Called the significance level
The probability of rejecting the null hypothesis when it is true.
Usu alpha= 0.05. It ranges from 0-1
If you are making airplanes: make that alpha low; if you are making paper airplanes: make that alpha high

28
Q

When is something statistically significant?

A

When the result of the hypothesis test is a p value less than 0.05

29
Q

T/F P value is the probability that your null hypothesis is true.

30
Q

If the p value is greater than alpha…what does that mean for the null hypothesis?

A

We fail to reject the null hypothesis. : (

31
Q

What is type I error? What represents this error?

A

Rejecting the null hypothesis when it is true.

alpha represents this.

31
Q

What is type I error? What represents this error?

A

Rejecting the null hypothesis when it is true.

alpha represents this.

32
Q

What is type II error?

A

failing to reject the null when it was false.

32
Q

What is type II error?

A

failing to reject the null when it was false.

33
Q

If you want a really awesome test…you choose a small/large alpha & make Type I/II errors more common.

A

small alpha
make it hard to reject the null
make Type II errors more common

33
Q

If you want a really awesome test…you choose a small/large alpha & make Type I/II errors more common.

A

small alpha
make it hard to reject the null
make Type II errors more common