Psych 505: Advanced Statistical Methods

Fall, 2000

Dr. David Carpenter, Ph.D.

 TEXT: Ferguson, G. A., and  Takane, Y. (1989) Statistical Analysis in Psychology & Education  (6th ed.)

GRADING: Grades will be based completely on the four 100 point exams and the special assignment #1 (50 points), except where I must make judgment calls. In such cases, in-class performance (i.e. preparation, participation, questions and, in general, your contribution) will count. Assignments must be handed in and will be checked over and returned, but will not be graded. Exams will be scheduled outside of class time on or about the dates indicated.

DATE

TOPIC

ASSIGNMENT

Aug. 29, 31,
 Sept. 5

Limits of class intervals, frequency distributions and polygons and their properties, averages, measures of variation, standard scores

Ch. 1: Ch. 2; pr 1, 2, 3, 5 & 6 (for Exercise 1
           only)
Ch. 3; pr 4, 5, 8, 9, 10
Ch. 4; pr 1, 2, 5, 9, 10
Ch. 5; (read 5.10 – 5.12 for general
           understanding.  You will not be  
           expected to do any calculations
           related to these sections.); pr 1, 2, 8

Sept.  7, 12

Probability and Binomial Distribution

Ch. 6; pr 1, 6, 11, 16, 18, 19, 20

Sept. 14

Normal Curve

Ch. 7; pr 3, 4, 8, 9

Sept.  19

Correlation and Regression

Ch. 8; (except 8.10) pr 3, 4, 8, 11, 13, 16, 17,
           22, 23
Ch. 23, pr. 5

Sept.  20 (Approx.)

EXAM I (Outside Class Time)

 

Sept.  21, 26

Sampling, sampling distributions, confidence intervals, t-distribution, degrees of freedom

Ch. 9; pr 5, 6, 9, 10
Ch. 10; pr 2, 5, 8, 12

Sept. 28, Oct. 3

Tests of Significance - means

Ch. 11; pr 1, 3, 7

Oct. 5, 12

Tests of Significance - variance, correlation coefficients

Ch. 12; (except 12.2, 12.3); pr 5-9

Oct. 17, 19

Analyzing frequencies with chi-square

Ch. 13; pr 1, 2, 4, 6, 7

Oct. 20 (Approx.)

EXAM II (Outside Class Time)

 

Oct. 24, 26, 31

Structure and planning of experiments, types of variables, one-way analysis of variance

Ch. 14; pr 4, 5, 7
Ch. 15; pr 2, 3, 7

Nov. 2, 7, 9

Two- and three-way analysis of variance

Ch. 16 (except 16.13 on); pr 1, 2
Special Assign. #2
Ch. 17; pr 2, 5

Nov. 10 (Approx.)

EXAM III (Outside Class Time)

 

Nov. 14, 16

Multiple comparisons

Ch. 18 (except 18.8-18.12); pr 1, 6, 7

Nov. 28,  30,
 Dec. 5

Some Repeated Measures Designs

Ch. 19 (except 19.11 to end of the Ch.);
            pr 2, 5

Dec.  5

SPECIAL ASSIGNMENT #1 DUE

 

Dec. 7, 12

Introduction to Multivariate Methods

Ch. 26; pr 1, 2, 3

Dec. 18  - 1:10 P.M.

EXAM IV (FINAL)

 

SPECIAL ASSIGNMENT #1: You are to examine the research literature and find a study using one of the techniques that we will study (preferably an ANOVA, but not necessarily so). This may be from a research report of just one study, or it may be one of several different studies reported in a given journal article. This assignment is due Dec. 2 nd and is worth 50 points. You will submit a copy of the article plus a typed (or printed) report providing the following:

I.      A full reference for the article.

II.     A description of the design and method

               A.     What is(are) the hypothesis(ses) in the study?

               B.     What is(are) the independent variable(s) and the dependent variable? How are they operationalized?

1.     If there is more than one IV, how are they combined ( factorially or otherwise)?

2.     Are the IV's manipulated or assigned?

3.     For each IV, is it between or within subject?

C.       How many total subjects were there in the experiment? How many in each treatment combination?
  1. Critique of the statistical analysis

A.      What level of scaling (nominal, ordinal, interval, ratio) is used for the dependent variable? What are the
           units of measurement?

B.     What statistical analysis is appropriate for this level of scaling and design?  If  an ANOVA, is it fixed,
         random, or mixed, and which variables are what?   Reconstruct the ANOVA Summary Table showing the
       sources of variation,  degrees of freedom, expected mean squares, and appropriate error term for each   F
         test.

C.     Try to figure out from the  F tests reported and the degrees of freedom, what  analysis was actually
         performed. Was it appropriate and correct?

D.     What assumptions are necessary for this analysis, and were they met?

E.     Were any additional (planned or post hoc) analyses done? What additional  assumptions do they require,
         and were they met?

F.     What do the statistically significant interactions look like? Draw a graph to show their form.

IV.      Conclusions

A.     What are the conclusions, and are they supported by the statistical analyses?

B.      Would other analyses that were not done be appropriate to do to test the  hypothesis(ses)?

SUGGESTED FOR GENERAL REFERENCE:

Barker, H. R., and Barker, B. M. (1984). Multivariate Analysis of Variance (MANOVA): A Practical Guide to Its Use in
          Scientific Decision Making. University, AL: University of Alabama Press.  A readable introduction to MANOVA, and
          a good place to start.

Cohen, J. & Cohen, P. (1986). Applied Multiple Regression Analysis for the Behavioral Sciences, (2nd ed.). A detailed
          in-depth coverage of the topic; especially good discussion of the uses of different codings.

Edwards, Allen L. (1985). Multiple Regression and the Analysis of Variance & Covariance, (2nd ed.). Relatively brief and
           readable coverage with good coverage of the relationship to ANOVA.

Edwards, Allen L. (1985). Experimental Design in Psychological Research (4th ed.) Good reference text, design oriented.

Grimm, Laurence G., and Yarnold, Paul R. (1995). Reading and Understanding Multivariate Statistics. Washington, D.C.,
           American Psychological Association. A non-computationally oriented explanation of multivariate statistics designed to
           explain the logic behind the various statistical methods.

Hays, William L. (1981). Statistics for the Social Sciences (3rd ed.). Theory oriented.

Kepple, G. (1991). Design and Analysis: A Researcher's Handbook (3rd ed.). Good reference, readable, complete, pragmatic
            answers to controversial issues.

Winer, B. J., Brown, D. R., and Michels, K. M. (1991). Statistical Principles in Experimental Design (3rd ed.). The bible, 
             quite extensive and detailed. Not easy reading.

Click here to return to Carpenter’s home page.

Hit Counter Since 8/22/00