Statistical Foundations with Excel
Kirit P. Mehta School of Law
Semester I & III — B.A., LL.B. (Hons.) & B.B.A., LL.B. (Hons.)
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 1 / 32
Course Overview
Duration: 45 Hours
Teaching Mode: Discussions + Excel Practice
Evaluation: ICA (100 Marks)
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 2 / 32
Course Objectives
1 Interpret and analyze data sets using spreadsheets
2 Solve problems with statistical tools in Excel
3 Make informed decisions using data-driven approaches
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 3 / 32
Unit 1–3: Excel Basics
Topics:
Excel interface and shortcuts
Functions: SUM, AVERAGE, IF, VLOOKUP, etc.
Pivot tables
Method: Hands-on using datasets emailed prior
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 4 / 32
Unit 4: Types of Data
Topics:
Qualitative vs Quantitative
Nominal, Ordinal, Interval, Ratio
Discrete vs Continuous
Tool: Excel with real-world legal data examples
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 5 / 32
Unit 5: Data Visualization
Topics:
Histograms, Scatterplots, Bar Charts
Shapes of Distributions
Book: Keller, Ch. 3
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 6 / 32
Unit 6: Sampling Methods
Topics:
Survey, Experimentation, Observation
Simple, Stratified, Cluster Sampling
Excel Practice: Sample generation and error identification
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 7 / 32
Unit 7: Central Tendency & Dispersion
Topics:
Mean, Median, Mode, Quartiles
Range, Variance, SD, IQR
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 8 / 32
Unit 8–9: Probability
Topics:
Sample Space, Events, Addition/Multiplication Rules
Conditional, Marginal, Joint Probability
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 9 / 32
Unit 10: Probability Distributions
Distributions:
Binomial
Poisson
Normal
Tool: Use Excel’s [Link], [Link]
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 10 / 32
Unit 11–12: Estimation and Hypothesis Testing
Concepts:
Confidence Intervals, p-values
Null vs Alternate Hypothesis
Type I and Type II Errors
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 11 / 32
Unit 13: Correlation
Topics:
Pearson and Spearman methods
Excel’s CORREL() function
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 12 / 32
Unit 14–15: Regression & Modeling
Concepts:
Simple and Multiple Linear Regression
Slope, Intercept, R 2 , ANOVA Table
Excel Output Interpretation: Coefficients, SE, F-test
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 13 / 32
Assessment Structure
ICA (100 Marks):
Quiz 1: 15
Quiz 2: 15
Presentation/Viva: 30
Assignment/Dissertation: 40
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 14 / 32
Textbooks and References
Primary:
Gerald Keller – Statistics for Management and Economics
Anderson et al. – Statistics for Business & Economics
Secondary:
Levin Rubin – Statistics for Management
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 15 / 32
Conclusion
Statistical thinking + Excel skills = Real-world advantage
All the best for the semester!
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 16 / 32
What is Statistics?
Simple Definition
Statistics is a way to get information from data.
That’s it!
Most of this course describes how, when, and why managers and
statistics practitioners conduct statistical procedures
Students will be exposed to different kinds of information and data
We’ll demonstrate these concepts through cases and examples
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 17 / 32
Statistics Practitioners vs. Statisticians
Statistics Practitioner
A person who uses statistical techniques properly
Examples of Statistics Practitioners:
Financial analyst developing stock portfolios based on historical rates
of return
Economist using statistical models to explain and predict inflation,
unemployment, GDP changes
Market researcher surveying consumers and converting responses into
useful information
Statistician
Individual who works with the mathematics of statistics, developing
techniques and concepts
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 18 / 32
Two Main Branches of Statistics
1. Descriptive Statistics
Definition
Methods of organizing, summarizing, and presenting data in a convenient
and informative way
Graphical Techniques:
Numerical Techniques:
Present data in ways that make
Calculate averages or means
it easy to extract useful
information Numerical statistical measures
(Chapter 4)
Variety of graphical methods
(Chapters 2 & 3) Describe different features of
data
Histograms for data distribution
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 19 / 32
Two Main Branches of Statistics
Example 3.3: Business Statistics Marks
Scenario: Student in business program asks professor about last year’s
statistics course marks
What information can be extracted?
1 Central Location - ”typical” mark
Average (mean)
Median
2 Variability - how scattered the marks are
Range (largest - smallest)
Standard deviation
3 Distribution - how marks are distributed
Best shown with histogram
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 20 / 32
Two Main Branches of Statistics
2. Inferential Statistics
Definition
A body of methods used to draw conclusions or inferences about
characteristics of populations based on sample data
We’re not interested in the sample itself
We want to know about the entire population
Interviewing entire population is prohibitive and time-consuming
Statistical techniques make large surveys unnecessary
Sample much smaller numbers and infer population characteristics
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 21 / 32
Case Study: Pepsi’s Exclusivity Agreement
Case 12.1: Pepsi’s University Deal
Situation: University offers Pepsi exclusivity agreement
50,000 students total enrollment
University gets 35% of revenues + $200,000/year
Pepsi has 2 weeks to decide
Current Pepsi Sales:
22,000 cans per week
40 weeks per year
$1.00 per can
30 cents cost per can
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 22 / 32
Case Study: Pepsi’s Exclusivity Agreement
Pepsi’s Analysis
If current market share is 25%:
Total market = 22, 000 ÷ 0.25 = 88, 000 cans/week (1)
Annual cans = 88, 000 × 40 = 3, 520, 000 (2)
Gross revenue = 3, 520, 000 × $1.00 = $3, 520, 000 (3)
After university take = 65% × $3, 520, 000 = $2, 288, 000 (4)
Costs = 3, 520, 000 × $0.30 = $1, 056, 000 (5)
Net profit = $2, 288, 000 − $1, 056, 000 − $200, 000 (6)
= $1, 032, 000 (7)
Current annual profit: $616,000 Potential gain: $416,000
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 23 / 32
Case Study: Pepsi’s Exclusivity Agreement
The Problem
Missing Information
Pepsi doesn’t know the total soft drink consumption at the university
Solution: Survey 500 students to track soft drink purchases for 7 days
This is inferential statistics:
Population: All 50,000 students
Sample: 500 students surveyed
Goal: Estimate total soft drink consumption
Data: Number of cans consumed in 7 days (file C12-01)
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 24 / 32
Example: Exit Polls
Example 12.5: Exit Polls
Scenario: 2000 Florida Presidential Election
Networks compete to predict winner first
Use exit polls - random sample of voters
Calculate sample proportion supporting candidates
Apply statistical techniques to determine if there’s enough evidence
Data: 765 people who voted for Bush or Gore (file Xm12-05)
Question: Can we conclude Bush will win Florida?
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 25 / 32
Example: Exit Polls
Exit Poll Analysis
Population: 5 million Floridians who voted for Bush or Gore
Sample: 765 randomly selected voters
Parameter of interest: Proportion of total electorate who voted for
Bush
Question: Did more than 50% vote for Bush?
Important Reality Check
We cannot predict with 100% certainty because we don’t ask all 5 million
voters. Statistics practitioners can control reliability and usually set it
between 90% and 99%.
Historical note: The networks made major errors that night, changing
predictions multiple times!
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 26 / 32
Key Statistical Concepts
1-1a Population
Definition
The group of all items of interest to a statistics practitioner
Frequently very large, may be infinitely large
Doesn’t necessarily refer to people (e.g., ball bearings from a plant)
Parameter: Descriptive measure of a population
Examples:
Case 12.1: 50,000 university students
Example 12.5: 5 million Florida voters
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 27 / 32
Key Statistical Concepts
1-1b Sample
Definition
A set of data drawn from the studied population
Statistic: Descriptive measure of a sample
We use statistics to make inferences about parameters
Examples:
Case 12.1: Mean soft drinks consumed by 500 students in sample
Example 12.5: Proportion of 765 Floridians who voted for Bush
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 28 / 32
Key Statistical Concepts
1-1c Statistical Inference
Definition
The process of making an estimate, prediction, or decision about a
population based on sample data
Why use samples?
Populations are almost always very large
Investigating each member would be impractical and expensive
Much easier and cheaper to sample
Reliability measures:
Confidence level: Proportion of times estimating procedure will be
correct
Significance level: Related measure of uncertainty
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 29 / 32
Types of Data
Data Types
Important Note
Data are not necessarily numbers!
Examples from our cases:
Student marks (Example 3.3): Numbers
Soft drinks consumed (Case 12.1): Numbers
Votes for Bush/Gore (Example 12.5): Not numbers (categories)
Different types of data require different statistical techniques, which we’ll
cover in Chapter 2.
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 30 / 32
Types of Data
Course Objectives
Our Goal
Convert you into a capable statistics practitioner who can use statistical
techniques properly
What we’ll cover:
Descriptive statistics (organizing and summarizing data)
Graphical techniques for data presentation
Numerical measures and their interpretation
Inferential statistics (drawing conclusions from samples)
Confidence levels and reliability measures
Practical applications using Excel
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 31 / 32
Types of Data
Questions?
Ready to explore how statistics helps us get information
from data?
Let’s begin with Excel!
Semester I & III — B.A., LL.B. (Hons.) & B
Kirit P. Mehta School of Law Statistical Foundations with Excel 32 / 32