Biostats

Question 1

Statistical significance only means the calculated p value is less than a present threshold

Accepted Answer

True

Question 2

dealing with CI, the larger the ________, the smaller the CI, which means we are _____ certain.

Accepted Answer

sample size, more

Question 3

P value of ___ is almost universally regarded as the cutoff level for statistical significance

Accepted Answer

0.05

Question 4

Significance level...Usually 0.05 – why not set much lower?

Accepted Answer

If you set too low, you will make many more Type II errors

Question 5

In regards to confidence intervals, for greatest assurance of the true mean, we want it to be..

Accepted Answer

Small

Question 6

with Survival Analysis, you want the starting time clearly defined, this should be an obj. date

Accepted Answer

Obtain documentation of first diagnosed

Question 7

You want the P-value to be lower, to show that the chance of obtaining extremes are _____.

Accepted Answer

Highly unlikely

Question 8

W/ Kaplan-Meier graph, each time a pt. dies the curve drops down. What do the dots represent?

Accepted Answer

Censored subject

Question 9

FDA specifies the entire __% CI of the ratio should be w/in that range (0.8-1.25, equiv to gen)

Accepted Answer

90

Question 10

If the wife says pie is better than cake, you're going to vote pie as well. This is problem of

Accepted Answer

Independent Observation

Question 11

Leading questions, such as, "Pie's better than cake right?" could skew that data. Problem of:

Accepted Answer

Accurate Date

Question 12

If give a drug, i.e. HTN Drug, you only expect the value, BP, to go in one direction. This is:

Accepted Answer

One-Tail P value

Question 13

•Survival with cancer is often quantified as ?-year survival

Accepted Answer

5

Question 14

Study is being done on group of ppl w/ HIV/AIDS...one person dies in a car crash. Must do what?

Accepted Answer

Don't remove observations, but report as censored

Question 15

If the Observed Mean and Null Hypothesis overlap, can you reject the null hypothesis?

Accepted Answer

No

Question 16

Can p values be negative?

Accepted Answer

No, always between 0-1.0

Question 17

One- or Two- Tail P Values are based on the same null hypotheses. True or False?

Accepted Answer

True

Question 18

previous data, physical limits, or common sense tells you the diff, can only go 1 direction use

Accepted Answer

one-tailed test

Question 19

Is concept of statistical hypothesis testing about making decisions or making conclusions?

Accepted Answer

Decisions

Question 20

When sample size is larger, the CI is_____

Accepted Answer

smaller

Question 21

What's better to compute...Mean survival time or Median survival time?

Accepted Answer

Median, b/c can be calculated before all the subjects die

Question 22

Normal body temp hypoth'd to be 37 C. if 95% CI range is 36.75 - 36.89 C,can reject hypothesis?

Accepted Answer

Yes, b/c significant difference is noted

Question 23

-survival time is recalculated w/ each pt. death – preferred method unless study is very large

Accepted Answer

Kaplan-Meier method (#alive at end of day/ # alive at begin)

Question 24

Confidence Interval depends on all EXCEPT what?

Accepted Answer

SEM

Question 25

To be more certain that an interval contains the true population value, you must generate a ___

Accepted Answer

Wider interval

Question 26

•A Type I error in statistics is the incorrect rejection of the true null hypothesis... aka

Accepted Answer

False positive

Question 27

Power can be used to do what?

Accepted Answer

experimental scatter (how messy is the basement), calculate the minimum effect size (how big the tool is), calculate the minimum sample size (time spent to find tool)

Question 28

When evaluating literature, do we want to look at the conclusion or the data?

Accepted Answer

data

Question 29

"probability that the test will reject the null hypothesis when the null hypothesis is false"

Accepted Answer

Probability of not making a Type II error, Power

Question 30

Most scientists are not so rigid & describe degrees of significance using a ______ scale

Accepted Answer

“Michelin Guide”

Question 31

If the observed mean does not overlap the null hypthesis, can you reject it w/out seeing the P?

Accepted Answer

Yes

Question 32

How can we get narrower CI's?

Accepted Answer

Improved methodology would reduce the SD (lower variability), Increasing the sample size in a repeat study

Biostats

Questions 32

Biostats

Questions 32