Unit – 1 Understanding Data

Introduction

In the field of Business Analytics, the first and most essential step is to understand the data. Businesses collect large volumes of information from sales, customers, operations, and markets. However, raw data alone is not meaningful. It needs to be imported, organized, cleaned, and summarized to generate insights that help in decision making.

This unit provides an introduction to the different types of data, their characteristics, and the common techniques used to represent and analyze them. The emphasis is on developing the ability to interpret and draw conclusions from data rather than performing technical coding.

Types of Data

Univariate Data

Univariate data consists of observations on a single variable.

Analysis of univariate data helps us identify patterns such as trend, seasonality, or overall performance level.

Multivariate Data

Multivariate data involves two or more variables observed for the same set of entities.

Analysis of multivariate data helps in understanding relationships and dependencies among variables.

Categorical vs. Quantitative Data

• Categorical Data: Represents labels or groups such as region, department, or product type. It cannot be measured on a numerical scale.

  Example: North, South, East, West (sales regions).

• Quantitative Data: Represents measurable quantities such as sales amount, profit margin, or number of customers. It can be expressed in numbers and subjected to arithmetic operations.

  Example: Monthly sales revenue in INR.

Data Preparation

Real-world data is rarely clean. It often contains errors, missing values, or inconsistencies. Before analysis, the following steps are essential:

• Importing Data: Bringing data into analysis software from sources such as spreadsheets, databases, or surveys.
• Cleaning Data: Correcting errors, handling missing values, and removing duplicates.
• Organizing Data: Structuring the data in tables for easy interpretation and analysis.

Sample Dataset

Let’s consider a small dataset from a retail company to understand these concepts:

Month	Region	Sales (INR)	Marketing Spend (INR)	Customer Satisfaction (1-5)
January	North	125,000	25,000	4.2
February	South	145,000	30,000	4.5
March	East	110,000	22,000	3.8
April	West	165,000	35,000	4.7
May	North	135,000	28,000	4.3

Descriptive Statistics

Descriptive statistics summarize data to make it easier to understand. Common measures include:

• Central Tendency: Mean, median, and mode which describe the typical or central value of the data.
• Dispersion: Variance, standard deviation, minimum, and maximum values which show the spread or variability in data.
• Frequency Distributions: Tables showing how often different categories or ranges occur in the dataset.

Descriptive Statistics for Our Sample Data

Measure	Sales (INR)	Marketing Spend (INR)	Customer Satisfaction
Mean	136,000	28,000	4.3
Median	135,000	28,000	4.3
Minimum	110,000	22,000	3.8
Maximum	165,000	35,000	4.7
Standard Deviation	19,364	4,472	0.33

Graphical Presentation of Data

Graphs and charts provide a visual summary of data, making it easier to identify patterns, relationships, and outliers. Common visual tools include:

Bar Plots

Used to compare frequencies or amounts across different categories.

Box Plots

Used to display the spread and distribution of data, highlighting the median, quartiles, and potential outliers.

Scatter Diagrams

Used to display the relationship between two quantitative variables.

Case Study Illustration

Consider a dataset from a retail company that includes three variables: Region (categorical), Sales (quantitative), and Marketing Spend (quantitative).

• A bar plot can show which region has the highest sales.
• A box plot can show how sales vary across regions and if there are extreme outliers.
• A scatter diagram can help in identifying whether higher marketing expenditure is associated with higher sales.
• Descriptive statistics can summarize the overall sales performance, average marketing expenditure, and variability across months.

From our sample data and visualizations, we can observe:

• The West region has the highest sales (INR 165,000)
• There appears to be a positive relationship between marketing spend and sales
• The East region shows lower sales despite moderate marketing expenditure, suggesting other factors may be at play