Inferential Statistics

Question 1

Inferential Statistics - Description

Answer 1

Refers to the generalization of results from a sample of participants to the whole population. Concluding weather a sample is significantly different from the population.
Hypothesis testing in general.

Question 2

Hypothesis Testing - Description

Answer 2

Determines the probability (p-value) of difference, or non-difference, between groups -> variable having an effect on the group. Comparing our info to the normal distribution. If they are outside of the 95% confidence interval, SD, then there is a difference.

Question 3

P-value

Answer 3

The smaller the p-value, the stronger the evidence is against the null hypothesis and in favor of the alternative hypothesis.
If it’s equal or inferior to 0.05, H0 is rejected in favor of Ha. The higher the p-value the more likelihood of random chance.

Question 4

Null Hypothesis vs Alternative Hypothesis

Answer 4

H0: always assumes the absence of difference. Example: Runners using more than one pair of shoes have the same risk of sustaining a RRI than runners who only use one pair of shoes.
Ha: assumes there is a difference. Example: Runners using more than one pair of shoes have a lower risk of sustaining a RRI than runners who only use one pair of shoes.

Question 5

Chi-Square Test (x2)

Answer 5

Used to determine if there is a significant association between 2 categorical variables (height, weight, gender…).
Compares the observed values with the expected.
You get P-value, if it’s below 0.05, then there is no significant association.

Question 6

Expected Values in Chi-square test - Formula

Answer 6

For example finding association between gender and BMI
Find expected values for each cell.
Cell is female, BMI 16-21: Total nr of female x total nr of BMI 16-21/Grand total.

Question 7

Student’s T-test - description, requirements and 2 types

Answer 7

How significant the group means are. Difference between the means of the groups and their corresponding variance.
Requirements:
- Continuous variables (scale)
- Normal distribution
- Equal variance in the samples
Paired samples T-test
Independent samples T-test

Question 8

Paired Samples T test

Answer 8

Two samples from the same group/individual.
Example: compare SLR values between lower limbs.

Question 9

Independent Samples T-test

Answer 9

Two samples from different groups/individuals. (one variable, two groups)
Example: compare SLR values between male and females.

Question 10

3 kinds of T-test

Answer 10

• Independent-samples: comparing the means of different groups.
• Dependent-samples: comparing means of 2 conditions where the same people are in both groups (paired sample t-test)
• One sample: comparing the mean of a sample with a pre-specified mean (comparing grades of this class with average grade of BPT12 in Lunex)

Question 11

3 main questions we should ask about the difference between groups?

Answer 11

1. Can I be certain that the difference is not due to random chance
2. How big is the difference?
3. Is this difference important?

Question 12

Student’s T-test - Formula

Answer 12

t = difference between group means/variability of groups
t = (avgA - avgB)/√S2A/NA+S2B/NB
S2=variance
S=standard deviation
N=sample size
avg=average
A=group A
B=group B
Taking into consideration that we have 2 groups so independent samples t-test.

Question 13

Student’s T-test - Calculating P-value

Answer 13

We compare the t-value to a threshold value corresponding to p<0.05.
t>threshold we assume Ha ->difference is significant. Low P = high T = Ha
t<threshold we assume H0 ->difference is not significant
T-value threshold= degrees of freedom = df=N-P
N=sample size
P=nr of variables