# What is the difference in interpretation of Pearson’s correlation and the coefficient of determination?

Question 1: 100 smokers are enrolled in a study to test the efficacy of a smoking-cessation drug. All participants are given the drug. The researchers contact the subjects at 3 months, 6 months, and 12 months; at each of those 3 intervals, the subjects are classified as either Smoking or Not Smoking. What statistical test should the researchers use to determine if there is a difference in the proportion of Smoking vs. Not Smoking at those 3 intervals? Explain your choice.

Question 2: How do you determine the degrees of freedom for Spearman’s rank order correlation?

**Question 3:**

What is the benefit of performing the Cochran’s Q test rather than examining frequency tables and performing McNemar testing alone?

**Question 4:**

What is the difference in interpretation of Pearson’s correlation and the coefficient of determination?

**Question 5:**

If you perform a Pearson correlation on multiple different continuous variables, do you have to run a Bonferroni correction on the corresponding p-values? Explain your reasoning.

**Question 6:**

Definition of Dichotomous data and 2 examples

**Question 7:**

McNemar test is:

a) The chi-square test of association for paired data

b) The Chi-squared test of association for ordinal data

C) The two way T-test for paired data

**Question 8:**

Why is that outliers are not acceptable in Pearson Correlation test? And if they are how we can correct them?

**Questions:**

**Question 9:**

Depending on the fact that McNemar test and paired t-test are used to determine if there is differences in the distribution between 2 related group, how can we decide which test to use?

**Question 9:**

What are the major assumptions that we need to check before running Pearson correlation?

**Question 10:**

When we should use Bonferroni adjustment of the critical alpha level?

**Question 11:**

What does pearson co-efficient measure? What is th range of pearson’s co-efficient?

**Question 12:**

A researcher attempts to understand if there is a linear relationship between two paired variables? What co-efficient should he use to do this?

**Question 13:**

A pharmaceutical researcher tries to understand the effects of two drugs and if the combination of the two drugs has a better effect than a single drug. The outcome of the drug use cannot be understood using a continuous variable so he decides to use as categorical variable. What non-parametric statistical analysis test should he use to carry out these analysis?

Question 14:

Let’s pretend. You are working on your final project and you realized you have to do non-parametric tests. Thinking of your final project for this class choose from these four test (Spearman’s Rank Order, Cochran’s IQ, McNemar’s Test and Pearson’s Correlation) which test would best represent your final project and data? Why did you choose this test? What is the null hypothesis? Does it meet the assumption? Will you have linearity or monotonicity? Outliers? Normality? And Why?

Question 15:

When testing for monotonicity, which test do you use when the scatterplot is in a linear direction? Which test do you use for a non-linear direction? Explain why?

Question 16:

Comparing Pearson’s Correlation and Spearman’s R, which tests is stronger and powerful? Explain why?

Q17. In the Pearson Correlation, what does an r value of -1 indicate?

Q18. True or false – in a Pearson Correlation between two variables, an r value of 1 indicates that one variable is causative of the variance in the other

Q19. A study is done with 100 participants to investigate the relationship between age and gender on alcohol consumption. Subjects recorded the number of drinks they consumed each week for 6 weeks. Subjects were categorized by gender and whether they were younger or older than 40 years of age. True or False — the McNemar Test is appropriate for determining whether there is a significant association between males and females under and over 40 and alcohol consumption. Why?