Lecture 12

Question 1

Paired samples

Answer 1

Independent observations of matched data.
Measure variable twice on the same subject ex before and after treatment.

Matched paired samples (age and sex)
*****More likely to find significance between subjects than if two randomly selected “unpaired” samples

Question 2

Unpaired samples

Answer 2

Measure a result once and compare between separate subjects.
Less likely to reveal a significant relationship
LARGER SAMPLE SIZE NEEDED WITH UNPAIRED THAN PAIRED.

Question 3

Paired data with tails

Answer 3

Two tailed test: More conservative approach. Ensures that either of two outcomes are covered.

One tailed test: Less robust. Assumes only one outcome likely. Rarely a safe assumption.

Question 4

Paired data with outliers

Answer 4

Extreme data points without obvious measurement induced errors.
Need common sense.

Question 5

Relative strength of relationship between variables top of pyramid to bottom

Answer 5

Top
Causal= predictors. Strongest. Parametric.
Correlation
Association
Random chance- not repeatable
Bottom

Question 6

Association is not necessarily

Answer 6

Causitive

Question 7

Most common test for correlation (association) and direction

Answer 7

Pearson’s product moment correlation coefficient (r)

Question 8

Assumptions for Pearson’s coefficient (r)

Answer 8

1. The population from which the sample is drawn is normally distributed. (If not, use non parametric test of correlation)
2. The two variables are structurally independent. (one not forced to vary with the other)
3. Only a single pair of measurements should be made on subjects.
4. Every r value (sample) should be accompanied by a p-value or confidence interval which the “true” R value (population) is likely to lie.

Question 9

For a parametric test, use the r value (sample) accompanied by a p value. For a non-parametric test, what coefficient do you use

Answer 9

r sub s for non-parametric instead of r

Question 10

perfect correlation value

Answer 10

1

Question 11

When to use Pearson’s r vs spearman’s rank

Answer 11

Pearsons: r. Normally distributed. Parametric.

Spearman’s: r sub s for non-normal distribution. Non parametric.

Question 12

Correlation does not allow for ___ or ___

Answer 12

Prediction or causal relationships. Only shows that there is a relationship between the two.

Question 13

How to interpret correlation

Answer 13

0. 9-1 very high positive correlation
   - 0.9 to -1 very high negative correlation
1. 7-0.9 high
2. 5-0.7 moderate positive/negative correlation
3. 3-0.5 low positive or negative correlation
4. 00-0.30 negligible correlation

Question 14

Regression analysis

Answer 14

Statistical modeling to estimate relationships among variables.
Used for prediction in parametric tests only.

Question 15

Linear regression

Answer 15

mathematical equation allows target variable to be predicted from the indecent variable. Continuous variables, linear relationship.

Slope intercept equation

Question 16

When to use the slope intercept equation

Answer 16

When determining linear regression

Question 17

Multiple regression

Answer 17

Not linear relationship between two or more independent (co-variables). May be quadratic or higher in nature.

Question 18

Probability and confidence is defined by

Answer 18

standard deviation. Defines probability limits

Question 19

SD probability limits

Answer 19

Approx 2 (1.96) SD above and below the mean defines points within which 95% of observations lie.

Question 20

Statistically significant P value

Answer 20

p < 0.05

Question 21

Statistically highly significant P value

Answer 21

p < 0.01

Question 22

Obtaining a significant or highly significant outcome means you should ___ the null

Answer 22

Reject.

Question 23

3 reasons you might fail to reject the null

Answer 23

1. No difference between groups
2. Too few subjects to demonstrate a difference existed. Small sample size.
3. Logical fallacy: Arbitrary assumption for cut off values. reality is values fall on a continuum.

Question 24

Confidence intervals allow for a ___ of response values in the form of a ____, given repetition of the study.

Answer 24

Confidence intervals allow for a continuum of response values in the form of a range of responses expected, given repetition of the study.

Question 25

The chance of a real difference given a CI lies between the

Answer 25

upper and lower limits. Difference is not statistically significant if either limit overlaps the value in question.

Question 26

CI can be applied to almost all statistical tests to help us understand if the evidence is:

Answer 26

Strong, weak, or definitive.

Question 27

Do you want a narrow or wide CI?

Answer 27

Narrow- greater precision.

Wider usually due to small sample size

Question 28

CI. Differences in populations. How do you know if there is no difference in populations?

Answer 28

The CI crosses zero or contains zero

Question 29

CI. Differences in ratios. How do you know if there is no difference in ratios?

Answer 29

The interval contains one or crosses one.

Question 30

CI on left side of null value

Answer 30

Results show statistically significant decline

Question 31

CI on right side of null value

Answer 31

Results show statistically significant improvement.

Question 32

Relative risk RR

Answer 32

Risk in treatment group/risk in control group

Question 33

Relative risk reduction (RRR)

Answer 33

percentage by which the risk of adverse event is reduced in the experimental group compared with the control group.

(risk in controls - risk in experimental)/risk in controls

Ex: reduced the death rate by 20%

Question 34

Absolute risk reduction (ARR)

Answer 34

Absolute amount by which a negative outcome is reduced comparing experimental with control group.

Ex: over time. Produced absolute reduction in deaths of 3%. Increased survival rate from 84% to 87%

Question 35

NNT number needed to treat

Answer 35

Number of subjects who would need to be treated to prevent one adverse outcome.

Reciprocal of the ARR (absolute risk reduction)

Presented with a CI

Ex: 34 people needed to enroll to avoid one death

Question 36

1/ARR=

Answer 36

NNT number needed to treat

Question 37

Hawthrone effect (observer effect)

Answer 37

Participants know they are being observed.