PSYB07 |
||||||||||||||||||||||||||||||||||||||||||
Data Analysis in PsychologySummer 2012
Instructor: Dr. Douglas A. Bors Office: P0103, room 111Office hours: My office hours during the examination period will be on Wednesday August the 15th from 11:00 to 2:00 and on Wednesday the 22nd from 11:00 to 1:30.
There will be review for the final examination on Wednesday August 22nd from 2:00 to 5:00 pm in room MW120. The TAs will have office hours on the follwing dates: Tuesday August 14th from 12:00 to 6:00 pm in room SW563, Wednesday August 15th from 1:00 to 3:00 in room SW418.
Assignment 1: Think of the day care example used in class. In this case, there are two batters: one has the higher overall batting average whereas the other has the higher batting average against both left-handed and right-handed pitchers. Batting average is calculated as the proportion of hits over total number of at bats. Make up a numerical example of this situation. What makes possible this apparent contradiction?
Assignment 2. Your assignment is to reproduce the overheads (save the first) on the PSYB07 webpage link called "Properties of Estimators." Create a population of three numbers. Then analyse all possible samples of two. For all samples calculate variance using both n and n-1. Then repeat this analysis using the population mean for each caluculation, rather than the specific sample means. The question is, in the two series of analyses, which formula ( n or n-1) produces an unbiased estimator and why? Assignment 3. Create a bimodal data set. You will likely need to use 30 or more observations. Then construct from that data two frequency histograms. The first should reflect the bimodal nature of the data. The second histogram should make the data appear to be unimodal. 4. Assignment #4. First, discuss the similarities and differences between the Phi coefficient and the correlation coefficient. Next, create two data sets in which the X and Y variables have the same mean and variance (7 pairs of Xs and Ys in both cases). In the first case the correlation between Xand Y should be very weak and approach 0.0. In the second case the correlation should be very strong and approach 1.0.. Textbook: Statistcal Methods for Psychology (7th ed.) by David HowellMath Review Quiz I.Math Review Quiz II.
This course is designed to provide the student with the basic principles of data analysis for both descriptive and inferential statistics. In terms of descriptive statistics, our treatment will include measures of central tendency, measures of variability, regression, correlations, and graphic presentations. Regarding inferential statistics, our introduction will include Chi Square, t-tests, and Analysis of Variance (one-way designs). A working knowledge of elementary algebra is assumed.Grading: Your final grade in the course will be based on quizzes and assignments (10%), a mid-term examination (40%), and a final examination (50%). There will be at least three in-class quizzes and two assingments during the term. We will take your best four marks. The quizzes will be administered in class or tutotial without warning, so be prepared! The assignments and their due dates will be announced in class. The date for the mid-term will be posted and announced early in the term. The date for the final examination will be published by the registrar's office sometime during the term.Make-UpsMake-up quizzes and assingments are not given. If a test is missed, do not phone or e-mail your instructor or TA concerning missed exams. Make-up mid-terms will be given at 5:00 pm on the Tuesday of the week following the original date of the mid-term exam. On the date of the make-up, the location of the exam will be posted on the office door (S-638) of Dr. Bors. If the make-up is also missed, a grade for the mid-term will be assigned on the basis of the student's relative performance on the final examination. Make-ups for final examinations are controlled by UTSC policy and the registrar's office.Dates for Exams will be posted at the top of this page, once they have been scheduledTentative Course Outline
Some Overheads for Classroom LecturesHere are the instructions for downloading and printing the overheads.Step #1: Click on the link from the list that corresponds to the overhead you wish to view.Step #2: A window opens asking what you wish to do with the file. Choose "open" and then click OK.Step #3 Under the file tab, choose the PRINT option. Note that in the window that pops up there is a "PRINT WHAT?" field. If you choose not to print them as slides (the default), you might print them as "handouts", which will put several on a sinle page and still leave you space for writing notes.Formula SheetSummation : the rules of summation notationDaycare : an example of contradictions in analysesBasicConcepts : the basic concepts of the fieldScales : a numbers is not a numbersCentral Tendency : different types of averagesComposite MeansInterquarilte RangeMeasures of SpreadProperties of Estimators----- optional derivation of sample variance (n-1) as unbiasedDegrees of FreedomGraphs & DistributionsStandardized ScoresNormal DistributionREVIEW Question set #1Hypothesis testing: first lookProbabilityReview Question Set #2Binomial DistributionReview Question Set #3Chi SquareRegressionOptional link on Simultaneous EquationsCorrelationsReview: Questions #4t-testANOVA assumptionsApproaches to ANOVAIn class example overheadsTest of Significance for Regression and CorrelationReview Question #5Final Review Questions |
||||||||||||||||||||||||||||||||||||||||||