Statistics Chapter 9 (Hypthesis Testing) Flashcards
What’s the difference between scientific and statistical hypothesis
Scientific hypothesis is a statement on how the world works (light will help people with depression)
Statistical hypthesis involves a parameter (the chance for a girl is 4.92)
Explain a significance test
Involves two statistical hypthesis
- Null hypothesis (specifies a particular value for a population parameter)
- Alternate hypothesis (specifies an alternate range for values (p is larger than 4.92))
What is the p value
The p value is the probability for results outside of H0 even though H0 is true
You find it by finding the z value und applying it in appendix A
What are the 5 steps of a significance test
- Assumptions (
- Hypothesis (null and alternative hypothesis)
- Test statistic (for proportion or mean)
- P value (calculate right tail probabilities)
- Conclusion (p value < Level of significance –> reject)
How does the significance test correlate with the confidence interval
Conclusions from a two sided significance test will always agree with a conclusion drawn from a confidence interval
A 99% confidence interval corresponds to a 0.01 significance value
Explain the types of error and how often they happen
- Type 1 error (H0 is rejected even tho it is true)
Most of the time this would be the worst case that could happen - Type 2 error (H0 is accepted even tho it is false)
What does the p value actually say
It actually states the probability for a type 1 error
Why is the confidence interval better than significance test
Because the significance test merely indicates whether a parameter value could work
A confidence interval is better as it displays the entire seit of believable values
What is the difference between statistical and practical significance
Practical significance may be different as we are talking about statistics and not about parameters
