Week 10 Discussion Questions
1.
What is statistical significance? It’s a determination of the null hypothesis which prof that the results are due to chance only.
2.
Why is it “harder” to find a significant outcome (all other things being equal) when the research hypothesis is being tested at the 0.01 rather than the 0.05 level of significance?
Less room is left for mistakes and errors at the 0.01 level because the test is more rigorous.
3.
Discuss the general idea that just because two things are correlated, one does not necessarily cause the other. Provide an example other than
ice cream and drowning.
The correlation between two things doesn’t mean they affect each other. An example could be the rise in fast food sales and homeliness rate. Even though both rose together, that doesn’t mean that the rise
in fast food sales caused homeliness.
Research Question 1:
Is there a difference between smoking behaviour (non-smoker, current smoker, or past smoker) and sex (binary for this question: male or female *but I recognize sex is not binary)? We could use a chi-square test to determine if there is a significant association between sex and smoking behaviour. 1.
State the null and research hypotheses?
Null: gender and smoking behavior are independent of each other.
Research: gender and smoking behavior are correlated.
2.
Decide on a confidence level?
0.05 significance
3.
What is the p-value cut-off?
0.0033
4.
Why are we using a chi-square test? Explain.
For test independence to compare the expected and observed frequencies.
5.
Interpret the findings?
Gender and smoking behavior are correlated to each other.
6.
Do you accept or reject the null? Reject the null hypothesis.
*Note cases represent subjects, and each subject appears only once in the dataset. That is, each row represents an observation from a unique subject.
