OLS RegressionSteps to Conduct an OLS Regression Analysis in SPSS Standard Multiple Regression • Click Analyze, then Regression, then Linear • Move your dependent variable to the “Dependent” box on the right • Move your independent variables to the “Independent(s)” box on the right • Click Statistics Click on Collinearity Diagnostics, optional: Descriptives, optional: Confidence Intervals and set a confidence level (e.g., 95), optional: Part and Partial Correlations Click Continue • Click OK Sequential Multiple Regression (Block-Entry Regression) • Click Analyze, then Regression, then Linear • Move your dependent variable to the “Dependent” box on the right • Move your first set of independent variables to the “Independent(s)” box on the right • Click Next in the upper right beside “Independents(s)” • Move your second set of independent variables to the “Independent(s)” box on the right. Repeat the prior step and this step until you have entered each set of variables. • Click Statistics Click on R squared change and Collinearity Diagnostics, optional: Descriptives, optional: Confidence Intervals and set a confidence level (e.g., 95), optional: Part and Partial Correlations Click Continue • Click OK Statistical Multiple Regression (Stepwise Multiple Regression) • Click Analyze, then Regression, then Linear • Move your dependent variable to the “Dependent” box on the right • Move your independent variables to the “Independent(s)” box on the right • Change “Method:” (to the lower right of the “Independence(s)” box) to Stepwise • Click Statistics Click on R squared change and Collinearity Diagnostics, optional: Descriptives, optional: Confidence Intervals and set a confidence level (e.g., 95), optional: Part and Partial Correlations Click Continue • Click OK 1. For this problem, you will use data from the National Football League (football.sav). Conduct a standard multiple regression analysis testing the hypothesis that the number of games won in a season (games) is a function of season passing yards (ydpass), the percentage of rushing plays (pcrush), and the opponents’ season rushing yardage (opprush). Use an alpha of .05 to answer the following questions. a. What is the computed F statistic and what does this indicate? (2 points) b. What is the computed coefficient of determination? Interpret what this number indicates? (2 points) c. Is there a problem with multicollinearity in this analysis? Why or why not? (2 points) d. Which of the independent variables have significant unique relationships with games won? (2 points) e. How do you interpret the unstandardized regression coefficient for opponents’ rushing yardage and for percentage of rushing plays? (2 points) f. What does the t value and accompanying significance level for opponent’s rushing yardage indicate? (2 points) g. What is the final regression equation resulting from this analysis (use the unstandardized regression coefficients to write the regression equation)? (1 point) h. What substantive conclusions would you come to from this analysis? (2 points) 2. Using the same dataset from the National Football League (football.sav), conduct a block-entry regression analysis analyzing the number of games won in a season (games). Enter percentage of rushing plays (pcrush) in the first block, season passing yards (ydpass) in the second block, and opponent’s season rushing yards (opprush) in the third block. Note the increments of variance explained (R-square change) by each block of variables. Use an alpha of .05 to answer the following questions. a. Which variable explains the largest proportion of variance? (1 point) b. Are each of the increments in variance explained significantly different from 0? (1 point) c. What is the final regression equation (use unstandardized coefficients)? (1 point) Now, change the order of entry such that opponents' rushing yardage (opprush) is entered first, then percent rushing plays (pcrush), and finally passing yardage (ydpass). Again, note the increments of variance explained by each addition. d. Which variable now explains the largest proportion of variance? (1 point) e. Again, is each of the increments in variance explained significant? (1 point) f. What is the final regression equation (use unstandardized coefficients)? (1 point) g. Compare each of the resulting equations to each other and to the equation from #1. How do they differ? (2 points) h. What's the point being made??!!! (2 points) OLS Regression 1 Steps to Conduct an OLS Regression Analysis in SPSS Standard Multiple Regression • Click Analyze, then Regression, then Linear • Move your dependent variable to the “Dependent” box on the right • Move your independent variables to the “Independent(s)” box on the right • Click Statistics Click on Collinearity Diagnostics, optional: Descriptives, optional: Confidence Intervals and set a confidence level (e.g., 95), optional: Part and Partial Correlations Click Continue • Click OK Sequential Multiple Regression (Block-Entry Regression) • Click Analyze, then Regression, then Linear • Move your dependent variable to the “Dependent” box on the right • Move your first set of independent variables to the “Independent(s)” box on the right • Click Next in the upper right beside “Independents(s)” • Move your second set of independent variables to the “Independent(s)” box on the right. Repeat the prior step and this step until you have entered each set of variables. • Click Statistics Click on R squared change and Collinearity Diagnostics, optional: Descriptives, optional: Confidence Intervals and set a confidence level (e.g., 95), optional: Part and Partial Correlations Click Continue • Click OK Statistical Multiple Regression (Stepwise Multiple Regression) • Click Analyze, then Regression, then Linear • Move your dependent variable to the “Dependent” box on the right • Move your independent variables to the “Independent(s)” box on the right • Change “Method:” (to the lower right of the “Independence(s)” box) to Stepwise • Click Statistics Click on R squared change and Collinearity Diagnostics, optional: Descriptives, optional: Confidence Intervals and set a confidence level (e.g., 95), optional: Part and Partial Correlations Click Continue • Click OK