Statistics

Question 1

What is regarded as the gold standard of scientific evidence?

Answer 1

RCTs

Question 2

Population

Answer 2

The whole

ex: all pts with malnutrition

Question 3

Sample

Answer 3

Part of the whole

ex: computer randomized selection of malnourished patients

Question 4

p-value

Answer 4

How likely there is to be an actual difference between groups

ex: p < 0.05 means the likelihood the result is due to chance is less than 5%

Question 5

Adequate Power

Answer 5

A study with a power of >/= 80% is considered a good study

Ex: There is an >/= 80% chance of detecting a difference as statistically significant if a true difference exists

Question 6

Type 1 Error

Answer 6

The error caused by rejecting a null hypothesis when it’s true (false positive)

Analogy to a false alarm

The probability of committing a type I error is equal to the level of significance that was set for the hypothesis test. Therefore, if the level of significance is 0.05, there is a 5% chance a type I error may occur

Question 7

Type 2 Error

Answer 7

The error that occurs when the null hypothesis is accepted when it is not true (false negative)

Analogy to a missed detection

Example, a test for a disease may report a negative result when the patient is infected. This is a type II error because we accept the conclusion of the test as negative, even though it is incorrect.

Question 8

Reliability

Answer 8

Is the data consistent and repeatable?

Question 9

Validity

Answer 9

Is the data meaningful and useful?

Question 10

Stats for reliability

Test-retest reliability

Answer 10

A measure of reliability obtained by administering the same test twice over a period of time to a group of individuals

The scores from Time 1 and Time 2 can then be correlated in order to evaluate the test for stability over time.

Question 11

Stats for reliability

Examples of Test-retest reliability

Answer 11

• Cohen’s Kappa (k) - categorical variables
• Intraclass Correlation Coefficient - continuous variables
• Cronbach’s alpha (a)- internal consistency

Look for one of these if a measurement is being done by multiple or a single user, or many people are taking a multi-item questionnaire

Question 12

Demographics table & statistical significance

Answer 12

Don’t want groups to be statistically different
>0.05 p value is a good thing here. It means that it’s the intervention is the cause, not the demographics

Question 13

Effect Size

Answer 13

• Indicates the practical importance of the outcome
• The larger the effect size, the more impact on the population
• A small effect size means that the research has limited practical applications
• Independent of sample size

Question 14

Cohen’s D

Answer 14

Used to measure the effect size

Question 15

Odds Ratio vs Risk Ratio

Answer 15

Comparing 2 different groups

Question 16

Odds Ratio

Answer 16

The ratio of 2 odds

ex: if an odds ratio is 1.52, this means there is a 52% increase (1.52 = 152%) in the odds of meeting the protein goal in the IF group as compared to the CF group

Question 17

Relative Risk/Risk Ratio

Answer 17

The ratio of 2 probabilities

If >1: the probability of meeting 80% of protein target in IF group was higher than the CF group
If =1: IF no better than CF
If <1: CF higher probability than IF