BUS1b2: Introduction to Statistics
Brandeis International Business School
Module
Time: Monday, 5:00—6:20
Location: Sachar International Center 116
Stephen Fournier
781-736-3898 Office: Heller Room 368
Fournier@brandeis.edu
Hours: TBA
Overview
This 2-credit module introduces statistical thinking and fundamental analytical methods to
undergraduate business students with little or no prior statistics training. In this module we will
develop an approach to enable critical analysis of statistical evidence to inform business
decisions. This will be accomplished by formulating both a way to think about data along with
hands on use of a statistical analysis package (Stata). Topics include the development and use of
both descriptive and inferential statistics, continuous probability distributions, estimation, tests of
hypotheses, and regression modeling.
We will take advantage of two free on-line statistics textbooks for most of our readings: David M.
Lane’s online Hyperstat as well as Statsoft’s Electronic Textbook. Articles and additional
readings will also be supplied on-line.
Rationale for the Course
As a primary part of their work, managers constantly make decisions. Fundamentally, these
decisions are all based on weighing the relevant information available. However, information is
frequently derived from measures that are often quite difficult to make directly, thus introducing
some degree of randomness. Even with numerous direct measures, many observed phenomena
contain a random component (to a greater or lesser degree). In either case, randomness is
important to managers because it means that the consequences of an action may not be
predictable with certainty. Reducing randomness, understanding its size and potential impact in a
management situation, is a valuable area of knowledge to which the study of statistics can
contribute greatly.
Not all randomness is the same—it derives from myriad sources and can be accounted for and
even reduced in various ways. Because some randomness reflects a lack of information, receiving
more or better information can lessen that part of the randomness, an idea behind quality control.
Although some phenomena are random, many random phenomena follow probability
distributions that have systematic features. Different random processes have some well-studied
probability distributions; statistical knowledge offers managers insight into these matters.
Statistical models provide a way to estimate the true relationships we may theorize to exist; to
inform the decisions we need to make. By incorporating the concepts of randomness, our models
also include a relative degree of certainty related to results—a way for us to test if our results are
‘statistically significant’.
With randomness and uncertainty comes the very real possibility of error. We may determine that
a result is ‘statistically significant’ and yet still be wrong. Even here, statistics offers a systematic
Page 1 of 7
January 20, 2016
way to examine how that error may come about. The models we will learn to develop and the
measures we will learn to use can offer insight into how to treat the various sources of error, the
probabilities of each of those errors, and ways to decide how to minimize the impacts of those
errors.
Learning Goals: A good manager must be able to understand measurement information provided
and use that information in a variety of ways. Statistical analysis and model building are primary
tools in this process. Upon completion of this class students will be able to:

Identify, evaluate, modify and manipulate basic data

Utilize this data for descriptive purposes by generating, analyzing and presenting a
variety of univariate statistics (sum, count, minimum, maximum, mean, median, variance,
standard deviation, etc.)

Utilize this data for inferential purposes by generating, analyzing and presenting a variety
of bivariate statistics (correlations, t-tests, f-tests, chi-squares, etc.)

Utilize this data for inferential purposes by generating, analyzing and presenting simple
regression models
Of course it is crucial that the use, manipulation and presentation of the data is carried out in a
fair and transparent fashion. Ethical behavior dictates that this process is not used to obfuscate or
mislead but rather to clarify and explicate the situation at hand.
These goals will be achieved through the understanding of a number of Key Concepts that we
will focus on throughout this course:





Populations vs. samples
Descriptive vs. inferential statistics
Generating and using summary measures for continuous data
o Central tendency
o Spread
The role of probability for use in prediction and statistical inference
A look at the basic statistical tests that use these concepts
o Correlations
o T-tests (1 and 2 sample) and ANOVA (3 or more samples)
o Chi-square
o Simple linear regression
Prerequisites
There are no formal prerequisites for this module. Students should be familiar with Excel and
basic algebra. If you have had more than one prior course in statistics, this course may be too
elementary for you. See me if you are in doubt. The target audience for this course is those
students with little or no prior background, and their needs will have the highest priority in the
class.
Keeping Informed
We’ll make regular use of LATTE and a course mailing list (registering in the course
automatically adds you to both lists). All lecture notes, handouts, links and other supporting
materials will be available via LATTE, and any late-breaking news will reach you via the mailing
list. Please check your Brandeis email regularly to keep apprised of important course-related
announcements.
Page 2 of 7
January 20, 2016
Participation & Contributions
IBS is truly an international community of learners, each with something to contribute to the
enterprise. Each student in the class should regard participation as a chance to contribute to our
joint efforts and helping fellow students to learn. Moreover, because this module aims to build
both understanding and effective communication skills, class participation is important. Statistics
is not a spectator sport; you will learn by doing rather than by watching. Participation can take
many forms, and each student is expected to contribute actively, freely, and effectively to the
classroom experience by raising questions, demonstrating preparedness and proficiency in the
analysis of problems and cases, and explaining the implications of particular analyses in context.
To this end, class attendance is required, and students should use name cards.
Written Assignments
We will have weekly problem sets, a midterm (both in-class and take home: to be done
individually), and a take home final (to be done individually). All written material is due one full
day before the beginning of the next class unless otherwise specified. In your written work, the
clarity and correctness of your explanations is at least as important as the numerical correctness
of your analysis.
Grading: The grade will be based upon a number of factors which are weighted approximately
as: class participation (10%), problem sets (20%), midterm (30%), and the final (40%).
Provisions for Feedback: Feedback will be provided along a number of paths. The problem sets
offer a rich weekly source on your progress, as well as in-class interactions. The midterm and
final will also provide strong indications of your progress. I will also offer a minimum of two
hours a week of office time during which I will be available if you have any problems/concerns
about your work. As I spend much of my time online, e-mail questions are absolutely welcomed
and I will respond as quickly as possible. The TA is also available for individual or group work
and will provide me another channel for feedback from students. I am absolutely available to any
individual who would like to speak with me: just ask.
Academic honesty: You are expected to be familiar with and to follow the University’s policies
on academic integrity (http://www.brandeis.edu/svpse/academicintegrity). Instances of
alleged dishonesty will be forwarded to the Office of Campus Life for possible referral to the
Student Judicial System. Potential sanctions include failure in the course and suspension from the
University. If you have any questions about my expectations, please ask. Academic dishonesty
will not be tolerated.
Notice: If you have a documented disability on record at Brandeis University and require
accommodations, please bring it to the instructor’s attention prior to the second meeting of the
class.
Study Groups
Working with one or two partners is an excellent way to gain deep under-standing of this subject.
I encourage small groups to work on assignments, with a few caveats:
 Be sure that you are neither carrying nor being carried by the group; each member of the
group is entitled to learn.
 For problem sets, you may work alone or with as many as 3 partners, but each person
should submit their own problem set.
 Each group member retains the right to “go it alone.” Joining a group is not a marriage,
and you can leave a group at any time.
Page 3 of 7
January 20, 2016
Week 1
Week 2
Topics Covered
Problem Set
Variation is a fact of life; problems and
opportunities in business context
Data and data measurement; types of variables.
Distinction between descriptive and inferential
statistics. Univariate statistics: summary
measures; central tendency; dispersion.
Stata
Introduction to the use of Stata. Opening files.
Generating basic statistics.
Readings for Next Meeting
Chapters 1-4 Lane text. StatSoft Chapters ‘Basic
Concepts’ and ‘Elementary Statistics.’ Read
Statistical Primer Handout.
Problem set 1 (generate
simple summary
statistics by hand and in
Stata)
Topics Covered
Huge groups, small groups; categories and
numbers; summarizing data in populations and
samples.
Discussion of summary measures: populations vs.
samples. Introduction to covariance and
correlation
Stata
First look at generating and interpreting
covariance and correlation in Stata.
Readings for Next Meeting
Statistical Primer Handout sections on covariance
and correlation.
Week 3
Topics Covered
Two by two, here and there, now and then.
Comparing groups, looking at variation over
time.
Covariance and correlation. Concept of population
vs. sample. Basic graphing.
Stata
First look at graphics in Stata.
Readings for Next Meeting
Handouts on graphics and z-scores
Page 4 of 7
Problem set 2 (generate
simple summary
statistics in Stata)
January 20, 2016
Problem set 3 (generate
univariate and bivariate
statistics along with
basic graphics).
Week 4
Topics Covered
Standardization; comparing some variables to a
predictable standard
Generating and understanding Z scores. More
graphic output.
Stata
Further look at Stata graphics.
Readings for Next Meeting
Introduction to concept of sampling distributions.
StatSoft Chapters 5-6.
Week 5
Topics Covered
Standardization; comparing some variables to a
predictable standard: continued
Another look at Z scores. Discussion of other
distributions. Discuss the use of probability
distribution functions (PDF) versus cumulative
distribution functions (CDF).
Stata
Review of Stata commands to date and further
look at data in Stata.
Readings for Next Meeting
Introduction to concept of sampling distributions.
StatSoft Chapter 7.
Week 6
Problem set 5 (open
and save data from
Stata to Excel and from
Excel to Stata)
Topics Covered
Where do samples come from?
A look at web-based data sources.
In-Class Demonstrations
In-class demonstration of finding, downloading
and using national datasets. Look at Gapminder,
GSS and/or others per class preference.
Readings for Next Meeting
StatSoft Chapters 7-8.
Page 5 of 7
Problem set 4 (interpret
z-tests).
January 20, 2016
No assigned problem
set
Week 7
Topics Covered
Leaps of Logic and Faith; Introduction to
inference and significance
Introduction to the t-statistic and measures of
significance. Note—next week we will have an inclass midterm.
Stata
Introduction to t-tests.
Readings for Next Meeting
HyperStat Chapter 12 (Anova).
Week 8
Topics Covered
Touching base; the midterm
In-class midterm for first hour.
Extension of t-test from two means to ANOVA
for multiple means (further use of F-test).
Stata
Introduction to ANOVA and F-tests.
Readings for Next Meeting
Handout on chi-square.
Week 9
Problem set 8
(generating and
interpreting Chisquare
results)
Topics Covered
Midterm Review
More complex data; A look at causal modeling
First a review of in-class midterm. Hand back and
discuss take home papers.
Introduction to linear regression.
Stata
Introduction to the regression command and more
on Chisquare: additional statistics for crosstabs
Readings for Next Meeting
Handout on regression output (UCLA website).
Page 6 of 7
Problem set 7
(generating and
interpreting ANOVA
results)
Topics Covered
Crosstabulations; crosstabs and Chi-square
Generating and interpreting chi-square tests.
Stata
Introduction to the tabulate command and
Chisquare: additional statistics for crosstabs
Readings for Next Meeting
Simple Linear Regression Handout
Week 10
Problem set 6
(generating and
interpreting t-tests.)
January 20, 2016
Problem set 9 (generate
and interpret simple
OLS model).
Week 11
Topics Covered
Modeling as a way of life; a look at various
models commonly used in business practice
Discussion of regression output: understanding
the overall F-test; R-squared, Adjusted RSquared.
Stata
More on regression and final comments on
Chisquare: additional statistics for crosstabs.
Looking at common regression models: trend-line
analysis, regression as a marketing tool, etc.
Readings for Next Meeting
Reading the Stata manual: review of commands to
date from manual.
Week 12
Topics Covered
Pulling it all together; Regression as correlation
and t-test and ANOVA: How they all relate
Present detail on regression as it relates to t-tests,
ANOVA and Chisquare. Demonstrate use of
overall F-test for regression. Demonstrate use of ttest as it relates to coefficients in regression..
Stata
More on regression and final comments on
Chisquare: additional statistics for crosstabs.
Looking at common regression models: trend-line
analysis, regression as a marketing tool, etc.
Week 13
Topics Covered
Full review for Final
Page 7 of 7
Problem set 10 (one
more time: dataset with
descriptives, ttest,
ANOVA and OLS)
January 20, 2016
Problem set 11 (one
more time: dataset with
descriptives, ttest,
ANOVA and OLS with
focus on interpreting
ALL output)