TECHCRUSH DATA

ANALYTICS BOOTCAMP
Facilitator: Omolola Okebiorun

Disclaimer: This training

Disclaimer: material
This training belongs
material belongs toto techcrush
TechCrush and shouldn’t
and shouldn’t be shared be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
  Basic data analysis concepts
Last
 Overview of analytical tools
class

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Why statistics?
• One • Two •Three

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

  Basic Measures of Central Tendency (mean,
median, mode)
 Sample and Population
Agenda
 Measures of Spread (range, variance,
standard deviation)

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Measures of
central tendency

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

• Measures of central tendency are simple ways to describe what a
typical value (representative) or middle of a set of numbers.

• The most common way to summarize a numerical data set is to

describe where the center is.

• The center of a data set can be measured in different ways, and

the method chosen can greatly influence the conclusions people
make about the data. The measures of center are the mean (or
average), median and mode.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 The mean is often called the average or the expected
value in mathematics.
We calculate the mean by adding all our values
together and dividing by the number of values in our
dataset.
Mean

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
  The median is the middle value.

 The median splits our data so that 50% of our values are
lower and 50% are higher.

Median  If we have an odd number of observations, the median is

simply the number in the direct middle.

 If we have an even number of observations, the median is

the average of the two values in the middle.

 Its important to sort the observations(values) you have

first.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Mode • The mode is the most frequently observed value in our
dataset.
• There might be multiple modes for a particular dataset
or no mode at all.
• No mode: means same frequency for all observations.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Activity 1

Find the mean, median and

mode of the following data
points.

5, 7, 8, 10, 10, 12, 13, 15, 15, 20

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Population
and
Sample

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Population Sample
• A population means the entire • A sample is a small part of the
group of people or items you population that you actually
want to study. collect data from.

• It includes everyone or • It represents the population

everything that fits your study but is smaller and easier to
interest. study.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Types of sampling

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Random sampling
• This is the way of choosing people or
items from a group so that everyone has
an equal chance of being selected.
• Example: Imagine you have 100
customers and you want to survey 10 of
them.
• You write all 100 names on pieces of
paper, mix them well and pick 10 blindly.
• Each customer has the same chance of
being picked.
• Avoids bias, fair representation
• Result of analysis reliable
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Systematic sampling
• This is the way of choosing items
from a group in a regular pattern.
Example: Imagine you have 100
customers and you want to survey
10 of them.
• Pick the first item and then select
every 10th item e. g 3, 13, 23, 33,…
• Easy to use
• Saves time
• Ensures coverage of the population
• Caution: If there is hidden pattern in
data, sample may be biased
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Stratified sampling
• This is the way of selecting a sample where you
divide a population into smaller samples (strata)
and then take a sample from each group.
• Example: Imagine you have 100 customers (60
male and 40 females) and you want to survey 10
of them.
• You divide them by gender and sample
proportionally.
60
• Male : × 10 = 6
100
40
• Female : × 10 = 4
100
• Fair representation, reduces bias, gives better
result when groups are different.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Cluster sampling
• This is the way of choosing items from a
group when a population is large or
spread out.
Example: Imagine you have 100 customers
and you want to survey 10 of them.
• Divide the population into groups using
purchasing power (clusters)
• Randomly select few clusters
• Collect data from all customers in the
selected clusters
• Used when population is too large,
geographically spread out, if random
sampling is expensive.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 1. Applications of standard deviation in

your field.
After-class
Adventure 2. Why sampling? Drawbacks of

sampling techniques.

3. Explain the non-probability sampling

techniques

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Measures of spread
A B

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Variability/Spread is what the field of statistics is all about.
Results vary from individual to individual, from group to group,
from city to city, from moment to moment.
Variation always exists in a data set, regardless of which
characteristic you’re measuring, because not every individual will
have the same exact value for every characteristic you measure.
Without a measure of variability you can’t compare two data sets
effectively.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Same mean and median but different
amounts of variability.
The first data set has a very small
Find the mean and
amount of variability compared to the
median of the following
second.
data points.

199, 200, 201

0, 200, 400

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Standard deviation is a measure of how spread out or
scattered the numbers in a dataset are from the mean
(average). It tells how much the values differ from the
average.
Lets look at two sets of scores
Set A: 10, 10, 10, 10, 10
Standard Set B: 5, 7, 10, 13, 15
deviation • Both have the same mean
• In Set A, no deviation from mean.
• In Set B, numbers are far from the mean (more
variation).
Small standard deviation means data is tightly
packed.
Large standard deviation means data is spread out.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Calculate the standard deviation of 2,4,6.
Step 1: Find the mean.
Step 2: Subtract the mean from each observation.
Step 3: Square the difference.
Step 4: Add the squares.
Step 5: Divide by the number of observations (This is the variance).
Step 6: Take the square root of the answer. Step 4
Step 1 Step 2 Step 3 4 + 0 +4 = 8
Mean = (2 + 4 + 2 -4 = -2 -22 = 4
+ 6 )/3 = 12/3 4-4=0 02 = 0
=4 6–4=2 22 = 4 Step 5
8/3 = 2.67
Variance = 2.67 Step 6
Standard deviation = 1.63 2.67 = 1.63
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Minimum: The smallest number in the dataset.
Q1​: The value such that 25% of the data fall below.
Q2​: The value such that 50% of the data fall below.
Q3​: The value such that 75% of the data fall below.
Maximum: The largest value in the dataset.

5 number
summary

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
  Outliers are points that fall
very far from the rest of our
data points.
 They influence measures
Outliers like the mean and standard
deviation much more than
measures associated with
the five-number summary.
 There are several different
techniques for identifying
outliers.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
When outliers are present, we should consider the following
points.
1. Noting they exist and the impact on summary statistics.

2. If typo - remove or fix.

3. Understanding why they exist, and the impact on questions

we are trying to answer about our data.

4. Reporting the 5 number summary values is often a better

indication than measures like the mean and standard deviation
when we have outliers.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 • Correlation is a statistical measure that describes
how two quantitative variables change in
relation to each other.

• It measures the strength and direction of a

relationship (association) between two
quantitative variables.
Correlation
• It is expressed as a value between -1 and +1
(correlationcoefficient).

• A positive correlation means both variables

increase together, while a negative correlation
means one increases as the other decreases.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 • Regression is a statistical method used to study
and predict the relationship between variables.

• It helps us understand how a dependent variable

(outcome) changes when one or more
independent variables (factors) change.
Regression • Dependent variable (Y): The variable being
predicted or explained. For example, a student’s
exam score.

• Independent variable (X): The variable used to

make predictions. For example, the number of
study hours.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Types of regression

• Linear Regression: Involves one

independent variable and one
dependent.

Example: Predicting exam scores (Y) from

study hours (X).

• Multiple Regression: Uses two or more

independent variables to predict one
dependent variable.

Example: Predicting sales (Y) based on

advertising, price, and income level.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Shape of
data
(Skewness
and
kurtosis)

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Histogram is the most popular visual for quantitative data.
From a histogram, we can quickly identify the shape of our data, which
helps influence all the measures we learned in previous concepts.
The above are the 3 main distributions frequently associated with
histograms. Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Normal
distribution

• Bell Curve.
Here the
•

mean =
median =
mode

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Right skewed

• Mean >
Median.
• Mode is
simply
the
longest
bin.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Left skewed

• Mean < Median.

• Mode is simply the
greatest bin.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Box plots for the histograms

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
Applications of statistical concepts
MEAN Median Mode
• Education: To • Economics: To • Retailers: The
calculate represent typical most frequently
average scores earnings (since a bought item.
to determine few very rich • Fashion: check
class people can raise the most
performance. the mean). common
• Business: • Real estate: clothing size.
Average sales Median house • Education: The
to evaluate prices to show most common
performance market trends grade among
over time. more accurately. students.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Applications of statistical concepts
Correlation
Standard deviation Regression
• Health: check
• Investors: Measure • Sales: Predicting
correlation between
risk — a stock with sales from
exercise and
a high standard advertising spend.
weight.
deviation has • Education:
• Economics:
unpredictable Estimating student
correlation between
returns. grades based on
inflation and
• Teachers: check study hours.
unemployment.
how much • Power supply:
• Marketing: explore
students’ scores Forecasting demand
correlation between
differ from the for electricity from
social media ads
class average. temperature
and customer visits.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Basic concepts of
Financial Mathematics

These are key financial and

business terms every data analyst
should understand, especially
when analyzing business or sales
data.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Profit

Profit is the money left after subtracting

all costs and expenses from total
revenue.
Profit=Revenue−Cost
Applications:
• Used to measure a company’s
financial performance.
• Profit trends help identify high-
performing products or regions.
• Helps decision-makers know whether
a business is growing or losing money.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Cost of Goods Sold (COGS)
COGS is the direct cost of producing or
buying the goods a company sells (e.g.,
materials, labor, packaging).

COGS = Opening Stock + Purchases −

Closing

Applications
• Used in profit analysis to determine gross
profit.
• Helps data analysts understand production
efficiency or supply chain costs.
• Important for pricing and budgeting
decisions.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Price
Price is the amount a customer
pays for a product or service.

Applications
• Price affects demand,
revenue, and profitability.
• Used in pricing models and
market analysis to find the
optimal price for maximum
profit.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Cost
Cost is the amount spent to
produce, purchase, or maintain
something.
Applications
•Data analysts compare cost
data to revenue to find profit
margins.
•Helps in budget forecasting
and expense control.

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

 Revenue
The total income a company earns from
selling goods or services before any
expenses are deducted.

Revenue=Price per Unit × Quantity Sold

Applications
• Key measure of business growth.
• Used to create sales dashboards and
forecast future income.
• Data analysts use revenue trends to identify
peak seasons or best-selling products.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Profit Before Tax (PBT)

PBT is the profit a business makes before

paying income tax.

PBT=Total Revenue−All Expenses (excluding

Applications
• Used to measure operational efficiency
before tax obligations.
• Analysts use it to compare companies
across regions or countries with different
tax rates.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 Discount
A discount is a reduction in price
given to encourage customers to
buy.
Applications
• Used in sales analysis to
measure how discounts affect
revenue and customer demand.
• Helps determine the optimal
discount rate that increases
sales without reducing profit
too much.
Disclaimer: This training material belongs to TechCrush and shouldn’t be shared
 What do the following mean?

After-class a. profitability ratio

Adventure
b. Liquidity ratio

c. Profit after tax

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared

Disclaimer: This training material belongs to TechCrush and shouldn’t be shared