METHODS OF ANALYSING DATA

Data analysis is a crucial process in understanding patterns, trends, and

insights that can inform decision-making in various fields. There are
several methods of analysing data, each suited to different types of data
and objectives. Below, I'll discuss the key methods of data analysis and
provide real-world examples of how each method can be applied.
1. Descriptive Analysis
Descriptive analysis involves summarizing and describing the main
features of a dataset. This method provides insights into past events and
is often the first step in data analysis. It uses measures of central
tendency such as mean, median, mode and measures of variability such
as range, and standard deviation, etc.
Additionally, it utilizes visualizations like bar charts, histograms, and pie
charts.
Real-world Example:
 Business Reporting: A company may use descriptive statistics to
summarize monthly sales data. By calculating the average sales,
identifying the most common sales values (mode), and creating
charts, the company can gain an understanding of their typical
performance over a period.
 Healthcare: Descriptive analysis can be used to summarize
patient demographics, such as the average age, most common
medical conditions, or the distribution of diseases in a particular
region.
2. Diagnostic Analysis
Diagnostic analysis is used to understand why something happened by
examining the causes or factors leading to a particular outcome. It goes
beyond simply describing data and seeks to identify patterns or
correlations that can explain an event.
Real-world Example:
 Retail: If sales drop during a specific period, a company may use
diagnostic analysis to investigate why. They might examine factors
like promotions, weather, customer behaviour, or competitors'
actions. This analysis can help determine the root cause of the
drop in sales.
 Healthcare: In an outbreak of a disease, diagnostic analysis might
be used to identify the factors (such as location, age group, or
specific behaviours) contributing to the higher incidence of the
disease.
3. Predictive Analysis
Predictive analysis uses historical data, statistical algorithms, and
machine learning techniques to predict future outcomes. It involves
building models that can forecast trends, behaviours, or events based on
existing data.
Real-world Example:
 Finance: Predictive analysis is commonly used in stock market
forecasting to predict future stock prices based on historical trends,
economic indicators, and company performance.
 Healthcare: Predictive models can be applied to patient data to
forecast the likelihood of a patient developing a particular disease
or condition. For instance, predictive analysis might help identify
patients at high risk of heart disease based on factors like age,
cholesterol levels, and smoking history.
 Retail: A retail business might use predictive analysis to forecast
future demand for products, allowing them to optimize inventory
levels and avoid stockouts.
4. Prescriptive Analysis
Prescriptive analysis recommends actions or decisions based on the
results of predictive analysis. It is considered a follow up from the
predictive analysis.
It suggests the best course of action to achieve desired outcomes and
often relies on optimization techniques, simulations, and decision
models.
Real-world Example:
 Logistics: A transportation company can use prescriptive analysis
to determine the most efficient routes for delivery trucks,
minimizing fuel costs and delivery times based on predictive traffic
data.
 Healthcare: A hospital can use prescriptive analysis to allocate
resources efficiently, such as deciding how many staff members
are needed during different times of the day or how to prioritize
patients based on the severity of their conditions.
 Marketing: A company can use prescriptive analysis to tailor its
marketing campaigns. By considering predictive models of
customer behaviour, it can recommend personalized marketing
strategies to maximize customer engagement and sales.

5. Exploratory Data Analysis (EDA)

Exploratory Data Analysis (EDA) is an approach to analysing data sets
to summarize their main characteristics, often with visual methods. EDA
is used to identify patterns, anomalies, or relationships in data, and it
forms the foundation for further analysis.
Real-world Example:
 Scientific Research: In genetics, EDA might be used to examine
the distribution of genetic variations across different populations.
Scientists would use boxplots, scatter plots, and correlation
matrices to identify any unusual genetic patterns.
 Customer Behaviour Analysis: A business might use EDA to
analyse customer transaction data to detect patterns in purchasing
behaviour. By visualizing customer spending habits and identifying
clusters of similar purchasing patterns, businesses can target their
marketing strategies more effectively.
6. Inferential Analysis
Inferential analysis involves using a sample of data to make
generalizations or predictions about a population. It relies on statistical
tests and probability theory to draw conclusions about a larger group
based on a subset of data.
Application to the real world
 Marketing Campaigns: Suppose a company wants to know
whether a new advertising campaign has increased sales. They
could collect a sample of sales data before and after the campaign
and use inferential statistics to determine if the observed change in
sales is statistically significant, thereby generalizing to the entire
population of sales.
 Election Polling: Polling organizations use inferential analysis to
predict election outcomes. By surveying a small, random sample of
voters, they can make inferences about the voting behaviour of the
entire population.
7. Text Analysis / Natural Language Processing (NLP)
Text analysis involves extracting useful information from unstructured
text data. Techniques like sentiment analysis, keyword extraction, and
topic modelling are used to derive insights from textual data.
Application to the real world
 Social Media Monitoring: Companies can use text analysis to
monitor social media mentions of their brand. Sentiment analysis
can be applied to determine whether customers’ posts are positive,
negative, or neutral, helping businesses understand public opinion.
 Customer Support: Text analysis can be used to analyse
customer support tickets and chat logs to identify frequently asked
questions, common issues, or emerging trends, helping
organizations improve their services.
8. Time Series Analysis
Time series analysis focuses on analysing data points collected or
recorded at specific time intervals. It is used to forecast future values
based on past patterns and trends, often with a focus on seasonality,
trends, and cycles.
Application to the real world
 Weather Forecasting: Meteorologists use time series analysis to
predict future weather conditions based on historical data,
including factors like temperature, precipitation, and wind speed.
 Stock Market Analysis: Investors often use time series analysis
to study stock prices and trends over time, helping them forecast
the future price movements of specific stocks or indices.
9. Cluster Analysis
Cluster analysis is an unsupervised learning technique used to group
similar data points into clusters. This method helps to identify distinct
segments or categories within a dataset based on shared
characteristics.
Application to the real world
 Market Segmentation: Businesses can use cluster analysis to
segment their customer base into groups with similar purchasing
behaviour, which can then be targeted with personalized marketing
campaigns.
 Healthcare: Cluster analysis could be used to categorize patients
with similar medical conditions, helping healthcare providers
identify common treatment patterns or high-risk groups.