Cumulative Final Assignments

1
Cumulative Final Assignments
2
MRGL34E1_5

• Assignment Due April 22 (Tu).
• MRGL34E1_5 - Please give this assignment name on
  your paper.
• Also, please "echo" each question on your paper
  prior to answering it.
• Exercises 1 5. Deals with aptitude scores
  predicting college stat course performance.
• Include a residual plot with your Results
  sections.
• Also include a Hierarchical MR Analysis Summary
  table modeled after Leech et al. class handout.

3
ParCorGL32E1_3

• Assignment Due April 22 (Tu).
• ParCorGL32E1_3 Name of assignment
• GS Lesson 32, Exs. 1-3
• When you answer specific questions, such as 1.a.,
  answer in a complete sentence rather than just
  giving the answer.
• For example, 1.a. The bivariate correlation
  between num_pubs and cites was r whatever.
• 1.b. Asks for a p value.
• Given the number of correlations computed, it
  would be good to apply the Bonferroni correction
  to get your critical p. However, 1b just asks for
  the p for num_pubs instructor quality
  correlation, not whether it is significant.
• Extra For 2, include a path diagrams for the
  instructor quality num_pubs relationship, with
  and without partialling out work ethic. Discuss
  whether the common cause hypothesis applies and
  to what extent.
• At the end as an appendix, include partial
  correlation outputs from both SAS and SPSS plus
  minimal/critical procedure syntax for each.

4
AOV1(GL25)

• Assignment Due April 17 (Th) by 5 PM.
• AOV1(GL25) - Please give this assignment name on
  your paper. Also, please "echo" the questions on
  your paper prior to answering it.
• 1. Create path diagrams for the two examples at
  the beginning of Green, Lesson 25.
•   Based on your study of Green, Lesson 25 and the
  class notes, create 2 path diagrams one each for
  the examples at the beginning of the chapter
  (p.183) -- the Experimental study example dealing
  with colds and vitamin C dose, and the
  quasi-experimental study dealing with student
  behavior problems and teaching philosophy. 
•   Note, in class I asked for 3 diagrams but am
  correcting that to 2 because the main example in
  the chapter is the same as the first example
  about vitamin C. 
• 2. What new or different assumptions are
  indicated for Anova compared to multiple
  regression. Explain briefly.

5
ACOV1(GL27)

• Assignment Due Monday, April 28 (by 5 PM) .
• Be sure to echo the question before answering
  it.
• 1. Create path diagram for the first example
  (same as main chapter example) on vitamin C and
  colds, Green L27. (2)
• 2. Read what Green has to say about the areas of
  application and the adequacy of the Ancova for
  each type of application. Note the example,
  given on p. 210, of a study with potential
  confounding. What is said about using Ancova in
  studies with potential confounding? Discuss the
  reasoning associated with this issue from an
  experimental design perspective. (2)
•  3. (a) What new or different main assumption is
  indicated for Ancova compared to Anova. Explain
  briefly. (2)
• (b) Assume a 2-group situation where Y is the
  DV, X is the IV, and C is the covariate. Make up
  sketches of two scatterplots with regression line
  (one for each group) illustrating a violation of
  the assumption. Make two more sketches
  illustrating the assumption being met. (2)
• 4. Use SPSS to create the graph shown in Fig.
  167, p. 219. Instructions are given in the first
  paragraph above on the same page. (2)
• 5. Read the data from Lesson 27 Data File 1.sav
  into SAS, run just the Ancova, also getting the
  adjusted means (lsmeans). Show /report output
  only that corresponds to the ANCOVA source table
  and Estimates, p. 216 in Green. No written
  report requested here. (2)
• ltassignment continues on next slidegt

6
ACOV1 Continued

• This next part is optional bonus for 4990
  students nonoptional for 6990 students.
• 6. (4 Points) Find the spss data file, Employee
  data.sav, found in the Program Files directory
  were spss is installed (file is one of many that
  comes with the spss program).
• The Employee data is from a bank in the Midwest
  during the late 60s that was involved in a sex
  discrimination EEOC probe.
• The assignment, basically, is to run an ANCOVA
  investigating gender differences in beginning
  salary (salbegin) controlling for education using
  SAS only. (Obviously, you can run spss to
  check/verify your results.) Gender will need to
  be recoded to numeric variable use SEX as the
  variable name.
• Turn in a Results Section (and SAS output for
  the Ancova and adjusted means as an appendix).
  For the Results Section, following the model from
  Green, p. 220, and
• Ignore and do not report (or test for) HOV of
  slopes. Just say, Homogeneity of slopes was
  assumed.
• Do report/include the statement on the
  covariate modeled after the one I put in the
  Ancova lecture ppt, slide 10.
• Your Results should include/address the question
  of whether there were gender differences in
  beginning salary, which gender, was
  disadvantaged, and whether and to what extent any
  differences were due to differences in education.
  
• This is an individual assignment. If you have
  questions or need a little help, contact me.
• Some SAS Help
• Data Bank
• Input id gender bdate mmddyy12. educ jobcat
  sal Dollar10. salbegin Dollar8. jobtime prevexp
  minority
• if gender 'm' then sex 1
• if gender 'f' then sex 0
• datalines
• ltinsert data here.gt