Hypothesis Testing
What is hypothesis testing used for?
Making decisions about whether your data support or refute your research predictions
What is a statistical error?
The unknown difference between the retained value and the true value
What is a Type I Error?
A false positive conclusion
What is a Type II Error?
A false negative conclusion
Not rejecting a null hypothesis when the alternative hypothesis is the true state
Accepting the null hypothesis when its false
Why is measuring statistical power important?
It is a value that shows the probability for correctly rejecting the null hypothesis
Predicting the research question: Test results shows Coronavirus is present and I do have Coronavirus
What is a null hypothesis?
The assumption of no difference between groups or no relationship between variables in the population
There is no relationship present
Always paired with an alternative hypothesis
How do you find a statistical error?
margin of error = critical value x standard deviation for the population
margin of error = critical value x standard error of the sample
What is an example of a Type I Error?(open for discussion)
Ex. Your test result for COVID-19 says you have coronavirus, when you actually don't
What is an example of a Type II Error? (open for discussion)
Ex. The COVID-19 test result says you do not have coronavirus, when you actually do
What factors affect statistical power?
Level of significance (alpha)
Effect size
Sample size
Statistical test we use
What is an alternative hypothesis?
Your research prediction of actual difference between groups or a true relationship between variables
What are two errors in statistics called?
Type I Error and Type II Error
What is a significance level? According to Type I Errors
This is a value you choose that you set at the beginning of your study to assess the statistical probability of obtaining your results (p value)
What decreases the chances of making type 2 errors?
Having a statistical power greater than 0.8 or 80%
The greater the power the less chance of making a type 2 error
There is an 80% of me correctly finding a statistical significance in my data
What are the different power tests that can used?
Priori analysis analysis
Post hoc power analysis
Why are errors unavoidable in hypothesis testing?
Making statistical decisions always involves UNCERTAINTIES
How do you control errors in Statistics? (Provide one answer for full points)
Ways to reduce measurement error:
double check all measurements for accuracy
double check your formulas are correct
make sure observers and measurement takers are well trained
make the measurement with the instrument that has the highest precision
make the measurements under controlled conditions
How do you reduce a Type I error probability?
Set a lower significance level
What can the researcher do in order to prevent a Type 2 error?
Get a large sample size!
What does the priori analysis test do?
Where do Type I and Type II Errors occur?
When the two distributions of null hypothesis and alternative hypothesis overlap
What causes errors? (provide one answer for full points)
sample size, sample design, sampling fraction, variability within the population
At what level of significance is it usually set in order to lower the occurrence of a Type I Error?
About 5%, 0.05, or a 5/100 chance
How is statistical power calculated?
1- B (remember B is the symbol of a Type 2 error)
What does the post hoc analysis test do?
It determines what power the study did occur
The sample size is already known