Explore BrainMass
Share

# Correlation and Regression

This content was COPIED from BrainMass.com - View the original, and get the already-completed solution here!

11A is a plot and regression analysis question from Norius Chapter 20, concept exercise # 4. I do not have this as a digital copy yet, so if you choose to accept this job, I will scan the problem and email.

11B
Using the Graphs menu, make a scatterplot of husband's education against wife's education (variables husbeduc and wifeduc). Edit the chart to draw the regression line and print R2
a. Does there appear to be a linear relationship between the two variables.
b. Would you characterize the relationship as positive or negative?
c. From the plot, estimate the slope and the intercept. (It's easier to estimate the intercept if you edit the x axis to start at 0.)
d. What's the value for the correlation coefficient?
e. Describe the points that are far from the regression line.

11C
From the Regression procedure, obtain the least-squares estimates for the slope and the intercept.
a. Write the regression equation to predict a husband's education from his wife's education. What proportion of the variability in husbands' education can be "explained" by wives' education?
b. What is the predicted value for a husband's education if his wife's education is 13 years?

11D
Again, this is from Norius chapter 21, Statistical Concepts exercise 3. I will scan and email the file -- a spss study looking for null and alternative hyotheses and analysis of that.

11E
Run a regression equation to predict father's education from mother's education (variables paeduc and maeduc). Include 95% confidence intervals for the slope and intercept. Save the standard error of the mean prediction
a. Write the linear regression equation to predict father's education from mother's education.
b. Based on the results of the linear regression, can you reject the null hypothesis that there is no linear relationship between father's and mother's education?
c. What proportion of the variability in mother's education is explained by father's education?
d. How can you tell from the slope if the correlation coefficient between the two variances is positive or negative.
e. What can you conclude about the population correlation coefficient based on what you know about the slope? Can you reject the null hypothesis that the population correlation coefficient is 0?

11F
Based on the regression equation developed in question 1, answer the following
a. What do you predict for father's education for a person who has a mother with 12 years of education?
b. What do you predict for average father's education for all people who have a mother with 12 years of education?

Add your analysis at the appropriate spots within the document, then save the file as a .rtf file using the NCU file naming convention and submit for credit.