Stat 1040, Section 002, Study Guide for Midterm 1 (Spring 2009)
Your first midterm exam is scheduled for Friday, February 13, 2009.
The exam is worth 200 points, i.e., 20% of your final grade.
A review session for Section 002 will be held
at the following place and time:
- Thursday February 12, 2009, 5:00pm-6:00pm, BUS 216.
Note that quizzes, midterms, and final exams from the most recent
semester can be found in the Stat 1040 workbook. Solutions for these quizzes
and exams and additional quizzes and exams from earlier semesters
can be found on the course Web site at
http://www.math.usu.edu/~symanzik/teaching/2009_stat1040/stat1040.html.
The first midterm will cover material from Chapters 1 through 10
of the Freedman, Pisani, and Purves book.
Midterm questions may originate from various sources, such as
questions from past midterms or past finals (as featured in the
Stat 1040 workbook or course Web page), review exercise
questions (assigned or non-assigned), past quizzes,
regular chapter exercises - and new questions!
The more questions (from various sources) you solve
before the midterm, the better you will be prepared!
However, just memorizing the answers from the Stat 1040
workbook or course Web page (for the old exams or for the review questions)
is not sufficient since numbers in the questions may
be changed (while the principle of the question remains unchanged).
Overall, there will be 4 to 6 questions in this exam.
You will be given about 50 minutes to complete the exam. The exam
will be a closed-book exam, which means that you are not allowed to
use the textbook or the lecture notes. All required tables (such
as for the normal curve) will be provided and required
formulas will be listed on the exam. You should bring your
calculator!
You will not be allowed to use your cell phone (either for
phone calls, text messaging, or as a calculator) during the exam.
Any violation of this policy will automatically result in a score
of zero for this exam.
To prepare for the exam, you should solve old midterm and final questions,
review exercise questions, questions from old quizzes,
and regular exercises from the textbook.
During the review session, it is
planned to discuss selected exercises from old exams.
Make sure that you are familiar with the keywords and concepts
listed below.
- Chapter 1 - Controlled Experiments:
randomized controlled experiment,
placebo, double blind, treatment group, control group,
treatment, response, historical controls
- Chapter 2 - Observational Studies:
association, causation,
confounding, controlling a confounding factor
- Chapter 3 - The Histogram:
drawing a histogram, comparing
areas of a histogram, density scale
- Chapter 4 - The Average and the Standard Deviation:
calculating average (mean), standard deviation (SD), median, and
interquartile range; effect of long tails on average,
68%-95%-99.7% rule, cross-sectional and longitudinal studies
- Chapter 5 - The Normal Approximation for Data:
normal curve, standard units,
finding area under the curve, normal approximation
for data, percentiles (and normal curve), change of scale
- Chapter 6 - Measurement Error:
chance error, outliers (and their treatment), bias (systematic error)
- Chapter 7 - Plotting Points and Lines:
[no direct question will be asked in the midterm -
however, knowledge of this chapter is required
for Chapters 8 through 10]
slope, rise, run, (y-) intercept, algebraic equation for a line,
plotting lines
- Chapter 8 - Correlation:
scatterplot (scatter diagram),
independent (explanatory) and dependent (response)
variables, correlation coefficient r,
computing r, guessing r, SD line, linear association,
strong/medium/weak positive/negative/no association,
perfect correlation
- Chapter 9 - More about Correlation:
features of r, linear association vs. association in
general (effect of non-linear association and outliers on r),
association vs. causation, ecological correlation
- Chapter 10 - Regression [as discussed by the end of the Wednesday lecture]:
regression line, constructing the regression line
(based on SD line and r), prediction, regression
effect, regression fallacy, two regression lines
Note that even if a keyword is only listed once, it may be related
to more than one chapter (e.g., confounding or SD).