|
|
|
|
|
|
|
|
|
|
PSY 769: Intermediate
Statistics in the Social Sciences
Spring 2014
"The primary purpose of the course is to
educate students about basic [statistical] theories and
techniques used in the behavioral sciences. The instructor
will briefly review information typically covered in
undergraduate statistics, and then introduce more advanced
statistical techniques. Upon completion, the student is
expected to understand the theoretical underpinnings for the
various statistical techniques and the assumptions that data
must meet to validly use these statistics. The student will
also gain an introduction to computer-based statistical
analysis. 30 hours. 3 credits." (from
Graduate Bulletin)
Many students enter the course several years
after having taken an undergraduate statistics course.
Moreover, new material often precedes review material in order
of logical development. As a consequence, we will
"briefly review" introductory material as if it were new
material and do so for most of the semester.
Nonetheless, from the start, we will cover most of the
material in more depth than a typical undergraduate course.
Computer based statistical analysis will
largely be limited to the use of spreadsheets. This is
because we offer another course entirely devoted to the topic
of computer based statistical analysis (PSY 737). I
highly recommend that course to both thesis and non-thesis
students.
Course
Objectives
1. Students will learn to view psychological phenomena from
the perspective of quantitative stochastic processes.
2. Students will develop linguistic competence in
interpreting, describing, and critically evaluating basic
statistical data.
3. Students will gain experience reasoning from and about
numerical data.
Note: It is not possible for this
course to cover all of the statistics that you might need for
a thesis. If you plan on doing a thesis, prepare
yourself for the fact that your data analysis will likely
involve at least some statistics not covered in this
course. See Blackboard for a document comparing what it
covered in this course to other benchmarks.
Required Reading
Bachman, R & Paternoster, R. (2008).
Statistical methods for criminology and criminal justice (3rd
ed.). New York: McGraw Hill.
Software
The course requires
the use of either Excel (available as part of
Microsoft Office in college computer labs) or Calc (freely available as part of
Apache OpenOffice at the following URL: http://www.openoffice.org/).
The programs are very similar, but there are some important differences.
Such spreadsheet software remains a popular
choice for data entry even
among researchers who use more specialized statistical
software for data analysis (the topic of PSY 737). However,
research shows that specialized statistical software can be more confusing
than enlightening
when learning fundamental statistical
concepts. Excel/Calc offer a simple computing
environment for working with data
ideal for learning and gaining confidence
with fundamental
statistical concepts.
Class Time
Do the reading before the corresponding class. If you
have questions about the reading, bring them to class and I will
try to answer them there. If you have trouble with the
homework, we can also go over that in class. I want to
leave some flexibility to use the class time in the manner you
will find most useful.
Homework
The homework assignment is the same for each assigned
chapter: (a) Provide a definition for each of the "Key
Terms" listed at the end of the chapter and (b) provide a brief
explanation of each of the "Key Formulas" listed at the end of
the chapter (note that most of these are in fact equations
rather than formulas). A good explanation paraphrases the
equation (or formula) in words, it tells the reader what the
formula says. Simply stating what the equation is used for
does not offer a good explanation of the equation itself.
Note that you do not fully understand an equation unless you
recognize each of the individual terms (e.g., n, s, x) used in
the equation.
Homework assignments are due at the start of
class for the day on which the reading was assigned. See
the course schedule at the end of the syllabus for
details. I recommend completing the homework as you read
the chapter rather than reading first and saving the homework
until afterward.
The examinations will not require calculations or software, but may include examples of spreadsheet formulas or results of spreadsheet calculations. The examinations will be multiple choice. Before the first examination, I will post a preparation guide to Blackboard.
At the final examination, you will have the
option of taking alternate versions of Test A and Test B to
possibly improve your grade on the earlier tests.
Grading
Your final grade comprises your three examination
grades (75%, 25% each) and your homework assignment grade (25%,
about 2% each). I
will return all grades in percent form, so to compute your
final grade you multiply by the above percents and add them
up: Final Grade = .25(Exam A) + .25(Exam B)
+ .25(Exam C) + .25(Homework Total). I
will assign letter grades as indicated below. Note that I will
do the above computation before rounding, so your results may
differ by rounding error in the reporting of individual
grades.
Grading on a curve normally refers
to forcing grades onto a normal distribution to fix the
proportion of students receiving each letter grade. I do
not grade on a curve because it makes it impossible for everyone
to do well. Anyone who meets the criteria can earn an "A"
in this course (or any other grade) no matter how other students
do. In other words, grading on a curve means that your
grade depends upon how well other students do in the
course. In this course, your grade only depends on how
well you do.
However, I will determine your grades using a
different kind of curve. The reason for this is that
multiple choice tests tend to be very effective assessment tools
in the sense that that they distinguish effectively between
different student's levels of understanding. That is, they
are very effective a spreading people out. However, a
variety of policies in the MA program were designed on the
assumption that MA course grades would have a compressed
distribution. To accommodate this assumption, I will
translate your proportion correct into a numeric course grade
using a nonlinear transformation. This transformation is
based on the an arc connecting the point <0,0> to the
point <1,1> along the edge of a circle with a center at
<1.5,-0.5> (as shown in the graph). Numeric Course
Grade (NCG) can be computed from your proportion correct grade
(PCG) using the following equation: NCG = 100 * (-0.5 +
sqrt(2.5 - ((PCG - 1.5)^2))). This amounts to a
diminishing returns curve such that each time Proportion Correct
increases, Numeric Course Grade increases by a little
less. As Proportion Correct increases, the proportion of
students with at least that Proportion Correct decreases, so the
result is the desired compression of the grade distribution
(squeezing students closer together).
I will use the following chart to convert Numeric Course Grades to Letter Course Grades. I will round x.5 and above up and anything below x.5 down.
Letter Grade | Numeric Course Grade | Approximate Percent Correct Grade |
A | 95-100 | 87-100 |
A- | 90-94 | 77-86 |
B+ | 85-89 | 68-76 |
B | 80-84 | 60-67 |
B- | 75-79 | 53-59 |
C+ | 70-74 | 47-52 |
C | 65-69 | 41-46 |
C- | 60-64 | 36-40 |
F | 0-59 | 0-36 |
Contact Information: (It usually works best to email me.)
Office
Hours: Tuesdays and Thursdays 4:15 PM to 5:15 PM.
Office: Room 10.63.11, 524 W59 Street.
Phone: 212-237-8784 (I do not check voice mail when off campus.)
Email: KMarkus@aol.com
Week |
Section 01 Tuesday |
Section 03 Thursday |
Assignments |
Topics |
1 |
1/28 |
1/30 |
Ch 1 (Recommended, Appendix A) No homework (HW) due. |
Why am I here? What is this class about?
Syllabus. Statistical inference and Sampling. |
2 |
2/4 |
2/6 |
Ch2-3 HW1: Ch 1. HW2: Ch 2-3. |
Levels of measurement, Distributions |
3 |
2/11 |
2/13 |
Ch 4-5 HW3: Ch 4-5 |
Central tendency, Dispersion |
4 |
2/18 |
2/27 (No Thursday Classes 2/20) |
Test A (Chapters 1-5) | |
5 |
2/25 |
3/6 |
Ch 6 HW4: Ch 6 |
Probability, Hypothesis Testing |
6 |
3/4 |
3/13 | Ch 7 HW5: Ch 7 |
Point estimation, Confidence Intervals |
7 |
3/11 |
3/20 |
Ch 8 HW6: Ch 8 |
Single group mean and proportion |
8 |
3/18 |
3/27 |
Ch 9 HW7: Ch 9 |
Hypotheses and Categorical data |
9 |
3/25 |
4/3 |
Ch 10 HW8: Ch 10 |
Two group mean and proportion |
10 |
4/1 |
4/10 |
Test B (Chapters 6-10) |
|
11 |
4/8 |
4/24 (Classes do not meet 4/17) |
Ch 11 HW9: Ch 11 |
ANOVA |
12 |
4/29 (Classes do not meet 4/15 or 4/22) |
5/1 |
Ch 12 HW10: Ch 12 |
Bivariate correlation and regression. |
13 |
5/6 |
5/8 |
Ch 13 HW11: Ch 13 |
Multiple regression |
14 |
5/13 |
5/15 |
Ch 14 (http://statpages.org/logistic.html) HW12: Ch 14 |
Logistic regression |
15 (Finals week) |
5/20 |
5/22 |
Test C (Chapters 11-14) |
|
|
|