# hw4 - mail 2 Read the article Bot et al Predictors of...

This preview shows page 1. Sign up to view the full content.

CR Long C532: Regression Modeling 30 Dec 2011 Homework #4 ( Due as e-mail attachments by 8am on Wed 7 Dec 2011 ): 1. Use the SPSS dataset HW4_C532 _2011.sav from the course website (20 pts). This dataset contains select data from a observation clinical study in patients with neck pain. Participants received 6 weeks of chiropractic care. Assessments were conducted at baseline and at the final visit during week 6. The primary outcome variable was the Neck Disability Index (NDI: 0 to 50 points, lower is better). We want to model the NDI at the week 6 visit with baseline data. a. complete an “All possible regressions” summary table b. select 2 possible “best” subsets of explanatory variables c. support your selection in b d. write out the fitted regression line for both models e. interpret both fitted models f. Would you pick one model over the other? Why or why not? Attach your summary table, written discussion and the SPSS output file to your e-
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: mail. 2. Read the article Bot et al. Predictors of outcome in neck and shoulder symptoms. Spine 2005;30:E459-E470 posted on the course website. (10 pts) a. Consider the second paragraph in the Statistical Analysis section. • Explain the authors’ process of arriving at their final regression models • Identify how the authors dealt with outcome variables that did not meet the model assumptions b. In Table 3 explain why all variables were not included in the models presented for both the 3 and 12 month outcome variable. c. For the 12 month outcome variable in Table 3 interpret the estimated beta coefficient for the variable “Duration of current episode.” d. For the 12 month outcome variable in Table 3 interpret the estimated beta coefficient for the variable “Fear-avoidance.”...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online