Unit 2 - Data Literacy

Data Pyramid is made of different stages of working with data

Moving up from the bottom
• Data is available in a raw form. Data in this form is not very useful.
• Data is processed to give us information about the world.
• Information about the world leads to knowledge of how things are
happening.
• Wisdom allows us to understand why things are happening in a
particular way.

How to become Data Literate?

Data Literate is a person who can interact with data to understand
the world around them.
A data literate person can –
• Filter the category as per the requirement – If the budget is low,
select the price filter as low to high
• Check the user ratings of the products
• Check for specific requirements in the product
🧠 Data Literacy Process Framework

🔹 1. Plan

What? Decide your goal and how long it will take.

Simple Example:
A school wants teachers to use student data to improve learning.
So, they plan to train teachers for 2 months.

🔹 2. Communicate

What? Tell everyone the plan and why it’s important.

Simple Example:
The school principal sends a message to teachers explaining why using data helps students
learn better.

🔹 3. Assess

What? Check what people already know.

Simple Example:
Teachers take a short quiz to see how comfortable they are with using data, like reading test
scores.

🔹 4. Develop Culture

What? Help people use data regularly.

Simple Example:
Teachers start looking at student scores every week to help students who are falling behind.
🔹 5. Prescriptive Learning

What? Give learning tools that fit each person’s style.

Simple Example:

• Some teachers watch short videos.

• Others join a live class.
• Some get printed guides.

🔹 6. Evaluate

What? Check if the plan is working.

Simple Example:
After 2 months, the school checks if teachers are using the data and if student scores are
improving.

What are Data Security and Privacy? How are they related to AI?

What is Data Privacy?

Data privacy, or information privacy, involves the proper handling of sensitive data—such as
personal, financial, and intellectual property—to meet legal requirements and ensure
confidentiality and integrity.
Examples

The following best practices can help you ensure data privacy:

Understand what data is collected, how it's used, and where it's stored. Collect only necessary
data and always prioritize user consent.

What is Data Security?

Data security is the practice of protecting digital information from unauthorized access,
corruption, or theft throughout its entire lifecycle.

Why is data privacy important?

With growing amounts of data stored in the cloud, the risk of cyber threats is increasing. It's
crucial to control and protect the transfer of sensitive information. Data security is more
important than ever because:
 • Cyber-attacks impact everyone
• Rapid technological growth is fueling more attacks

H.W

Best Practices for Cyber Security

Types of data
2.2.2 Data Acquisition
Data acquisition is the process of gathering data, often for training AI models. It includes
three main steps:

1. Data Discovery
o Involves searching for relevant datasets.
o Example: For a self-driving car CV model, collect images of roads and objects
from the internet.
2. Data Augmentation
o Expands existing data by making small changes (e.g., brightness, color).
o Helps increase dataset size without collecting new data.
3. Data Generation
o Involves recording new data using sensors.
o Example: Capturing temperature readings and storing them digitally.

H.W
Sources of data

• Primary data
• Secondary data
Ethical concerns in data acquisition
Features of data and Data Preprocessing

Usability of Data

There are three primary factors determining the usability of data:

1.Structure- Defines how data is stored.
For example, data stored in spreadsheet will have good structure
whereas the same data stored in text document will have poor
structure.

2.Cleanliness- Clean data is free from duplicates, missing values,

outliers, and other anomalies that may affect its reliability and
usefulness for analysis.
3.Accuracy- Accuracy indicates how well the data matches real-
world values, ensuring reliability. Accurate data closely reflects
actual values without errors, enhancing the quality and
trustworthiness of the dataset.

Features of Data

Data features are the characteristics or properties of the data. They

describe each piece of information in a dataset.
For example,
• In a table of student records, features could include things like
the student's name, age, or grade.
• In a photo dataset, features might be the colors present in each
image. These features help us understand and analyse the data.
In AI models, we need two types of features: independent and
dependent
Independent features are the input to the model—they're the
information we provide to make predictions.
Dependent features, on the other hand, are the outputs or results of
the model-they're what we're trying to predict.
In research, researchers often manipulate and measure independent
and dependent variables to test cause-and- effect relationships.

Data Processing and Data Interpretation

Data Processing
• Data processing helps computers understand raw data.
• Use of computers to perform different operations on data is
included under data processing.

Data Interpretation
▪ It is the process of making sense out of data that has been
processed.
▪ The interpretation of data helps us answer critical questions using
data.
Understanding some keywords related to Data
• Acquire Data- Acquiring data is to collect data from various
data sources.
• Data Processing- After raw data is collected, data is
processed to derive meaningful information from it.
• Data Analysis – Data analysis is to examine each component
of the data in order to draw conclusions.
• Data Interpretation – It is to be able to explain what these
findings/conclusions mean in a given context.
• Data Presentation- In this step, you select, organize, and
group ideas and evidence in a logical way.

Qualitative Data Interpretation

Types of Data Interpretation

How to interpret Data?
Based on the two types of data, there are two ways to interpret data-
• Qualitative Data Interpretation

• Quantitative Data Interpretation

 Qualitative Data Interpretation
Data Collection Methods – Qualitative Data Interpretation
Record keeping: This method uses existing reliable documents
and other similar sources of information as the data source. It is
similar to going to a library.
Observation: In this method, the participant – their behaviour
and emotions – are observed carefully
Case Studies: In this method, data is collected from case
studies.
Focus groups: In this method, data is collected from a group
discussion on relevant topic.
Longitudinal Studies: This data collection method is performed
on the same data source repeatedly over an extended period.
One-to-One Interviews: In this method, data is collected using
a one-to-one interview.
5 Steps to Qualitative Data Analysis
1. Collect Data
2. Organize
3. Set a code to the Data Collected
4. Analyse your data
5. Reporting
Quantitative Data Interpretation
▪ Quantitative data interpretation is made on numerical data
▪ It helps us answer questions like “when,” “how many,” and
“how often”
▪ For example – How many numbers of likes on the Instagram
post?

Data Collection Methods -Quantitative Data Interpretation

▪ Interviews: Quantitative interviews play a key role in collecting
information.
▪ Polls: A poll is a type of survey that asks simple questions to
respondents. Polls are usually limited to one question.
▪ Observations: Quantitative data can be collected through
observations in a particular time period
▪ Longitudinal Studies: A type of study conducted over a long
time
 ▪ Survey: Surveys can be conducted for a large number of people
to collect quantitative data.

4 Steps to Quantitative Data Analysis

1. Relate measurement scales with variables
2. Connect descriptive statistics with data
3. Decide a measurement scale
4. Represent data in an appropriate format

Types of Data Interpretation (DI)

There are three ways in which data can be presented:
• Textual
• Tabular
• Graphical

Textual Data Interpretation

▪ The data is mentioned in the text form, usually in a paragraph.
▪ Used when the data is not large and can be easily comprehended
by reading.
▪ Textual presentation is not suitable for large data.
Eg: "A company conducted a survey to find out employee satisfaction
levels. Out of 500 employees:300 said they were satisfied with their
job.120 said they were neutral.80 said they were dissatisfied.
Interpretation- 60% (300 out of 500) of employees are satisfied.
24% are neutral.
16% are dissatisfied.

Tabular Data Interpretation

• Data is represented systematically in the form of rows and
columns.
• Title of the Table (Sales Growth (2020 to 2022) contains the
description of the table content.
• Column Headings(Year; Product A Sales (in units); Product B
Sales (in units); contains the description of information contained
in columns.
Importance of Data Interpretation
• Informed Decision Making – A decision is only as good as the
knowledge is based on. Eg: Since the average height of students is
known, school can custom design the furniture according to the
requirement of the class.

• Reduced Cost – Identifying needs can lead to reduction in

cost. Eg: Based on the review or survey from customer, production of
the items with low review/interest can be drop/modify.
• Identifying Needs – We can identify needs of people by Data
Interpretation. Eg: Pizza is a popular choice among age group 8-10