Simple Regression Analysis Project
The Simple Regression Analysis Project is a required
assignment for STAT220: Introduction to Statistics. It represents 15% of the
student’s course grade. The project requires that students not only perform the
statistical procedures but also discuss the implication of these procedures to
the underlying problem. Students demonstrate not just the mere mechanics of
statistics, but understanding of the significance of statistical data. The
project may be an individual or group project.
• POSE THE QUESTION
Before you even start thinking of the relationship to
investigate you need to make sure you understand the concept of linear
relationship from MATH 125 well. Then
you should study the examples in the Course Notes and the Blackboard. Once you understand what kind of
relationships can be investigated using linear regression, then you need to
look for the availability of the data, and search for prior studies on this
relationship. During the process it will
be a good idea to confer with your instructors on the appropriateness of the
topic and the data.
• DATA COLLECTION
One component of the Simple Regression Analysis Project
involves data collection. Instructors may allow students to use existing data
sets or require students to gather data by developing a survey that they
create, on a topic of their choosing. Students may be encouraged to survey
their place of work, or choose a topic related to their majors, if possible.
Instructors also have the option of requiring that students obtain data from federal
or state government websites, i.e., census or labor data.
The data set should have the following properties:
• The sample size
should be at least 30 paired observations
• There are at
least two quantitative variables to be used for the simple linear regression.
Additional quantitative variables can be gathered if the student wishes to
investigate which independent variable may be a better predictor of the
dependent variable
Web Sites with Data Sets:
•
http://www.keycurriculum.com/resources/fathom-resources/free-activities-and-resources/statistics-activities
•
http://nilesonline.com/data/
•
http://www.dartmouth.edu/~chance/teaching_aids/data.html
•
http://lib.stat.cmu.edu/DASL/ and
http://lib.stat.cmu.edu/datasets/
• http://www.amstat.org/publications/jse/
•
http://www.fedstats.gov/
•
http://www.rossmanchance.com/ws2/mtw/index.htm
•
http://it.stlawu.edu/~rlock/datasurf.html#Individual
•
http://faculty.babson.edu/turner/fish.html
•
http://sunny.moorparkcollege.edu/~kfink/statlinks.htm
After collecting, enter the data into Minitab and send the
Minitab data set to the Instructor for the final approval.
• DATA ANALYSIS
The Simple Regression Analysis Project requires students to
perform each of the following:
• Use histograms,
box plots, dot plots, and descriptive statistics to study the center,
variation, distribution, outliers (if any), and trend (if applicable) for each
of the quantitative variables (students should decide not only whether the data
are symmetric or skewed, but whether the data are sufficiently symmetric to
make the assumption of normality)
• Use a scatter
plot, numeric measures of correlation, and
simple linear regression to examine whether evidence of a relationship
between two variables exists, how strong that relationship is, and whether the
regression equation should be used for predictive purposes
• Requirements of
written report:
• Three to five pages of written text, not
counting appropriate front and end matter, tables and graphics
• Typed,
double-spaced with 12-point font, one inch margins
• Formatted
according to the Publication Manual of the American Psychological Association
(APA).
• Free from
spelling, punctuation, or usage errors





