Statistics & Chemometrics Flashcards
Final prep
Three reasons sample results can differ?
- Actual differences
- Systematical errors
- Random variations
What is significance testing?
Mathematical method used to investigate if differences between samples can be explained purely by chance.
What is a null hypothesis?
The assumption that there is no significant difference between the groups being sampled.
(H0)
What is an alternate hypothesis?
The assumption that significant difference between the groups being sampled.
(HA)
H0: x1(average) = x2(average)
Null hypothesis
HA: x1(average) ≠ x2(average)
Alternate hypothesis
Can you accept H0 as true if there is not enough evidence to reject it?
No
What is a type I error?
A false positive, where H0 is rejected but it is actually true.
What is a type II error?
A false negative, where H0 is accepted but is actually false.
What does a confidence interval with 0 within the boundary imply?
H0: Δ(average) = 0
No significant difference between samples
What does a confidence interval with 0 outside the boundary imply?
H0: Δ(average) ≠ 0
Significant difference present between samples
What is a confidence interval?
The probability that a result will fall between a specified interval % of the time.
What is the likelyhood of getting a certain value within a confidence interval?
It is equal for all values within the interval.
Three factors that influence the size of a confidence interval?
- The confidence level
- Random variation present
- Number of samples
What is an independent test?
A test that determines whether there is a statistically significant difference between the means in two unrelated groups.
Example: Interviewing two groups about different teaching methods they were taught.
