Unit 2: Data, ethics and industry applications
Basics of Data Analytics using Spreadsheet notes · PTU syllabus (UGDSE101)
On this page
Unit summary
Analysis is only as good as the data behind it. This unit covers where data comes from, how it is collected, its formats, how to clean and transform it, how to handle missing values and outliers, the ethics of using data, and how analytics is applied in finance, marketing and operations.
After this unit you can
- Identify data sources, collection methods and formats
- Clean and transform raw data
- Handle missing data and outliers
- Explain the ethical issues in data analytics and give industry case studies
PTU syllabus topics
- Data collection methods
- data sources and formats
- data cleaning and transformation
- handling missing data and outliers
- ethical considerations in data analytics
- industry-specific applications (finance, marketing, operations) with case studies
Topic 1
Data collection methods, sources and formats
- Primary collection: surveys and questionnaires (Google Forms), interviews, observation, experiments, sensors.
- Secondary sources: company databases, government portals (data.gov.in), reports, websites, APIs.
- Common formats: CSV (comma-separated values), Excel (XLSX), JSON, XML, text files and databases.
Example
Primary and secondary: a Google Form sent to 200 students about canteen food gives primary data. Census figures downloaded from a government website are secondary data. Common formats are CSV, XLSX, JSON and XML.
Topic 2
Data cleaning and transformation
Cleaning usually takes the largest share of an analyst's time
Data cleaning fixes or removes incorrect, incomplete, duplicate or badly formatted data.
- 1
Remove duplicates
- 2
Fix formats
Dates, text case, units
- 3
Correct errors
Typos, impossible values
- 4
Handle missing values
- 5
Treat outliers
- 6
Transform
New columns, grouping, scaling
Transformation reshapes data for analysis: splitting a full name into first and last names, converting text to numbers, creating age groups, or calculating totals.
Example
Before and after: a raw sheet has "ludhiana", "Ludhiana " and "LUDHIANA" as three different cities, two identical rows and dates typed as text. After TRIM, PROPER, Remove Duplicates and converting text to dates, a PivotTable counts the city correctly.
Exam tip
Clean the data before analysing it. Most errors in results come from dirty data.
Topic 3
Handling missing data and outliers
Check whether an outlier is an error or a genuine extreme; options are to correct, remove, cap or keep it. Outliers pull the mean and distort the fitted line
| Problem | Options |
|---|---|
| Missing values | Delete the row (if few), fill with mean/median/mode, use a default value, or estimate from other columns |
| Outliers | Check whether it is an error; correct or remove errors; keep genuine extreme values but analyse them separately; cap them |
An outlier is a value far from the others. A common rule: values below Q1 − 1.5 × IQR or above Q3 + 1.5 × IQR are outliers (IQR = Q3 − Q1).
Exam tip
Never delete outliers blindly — a very large purchase might be your most important customer, not an error.
Example
The IQR rule: for marks with Q1 = 40 and Q3 = 70, IQR = 30. The limits are 40 − 1.5 × 30 = −5 and 70 + 1.5 × 30 = 115. Any value outside these is an outlier. A missing mark can be left blank, filled with the mean or median, or the row removed, depending on how many are missing.
Topic 4
Ethics in data analytics
- Privacy and consent: collect personal data only with permission and for a clear purpose (India's Digital Personal Data Protection Act, 2023).
- Security: protect data from leaks and unauthorised access.
- Bias and fairness: biased data leads to unfair decisions.
- Transparency and honesty: do not manipulate charts or cherry-pick data.
Example
An ethical lapse: an app collects users' contacts and location without telling them and sells the data to advertisers. Ethical practice asks for consent, collects only what is needed, hides names when sharing (anonymisation) and keeps the data secure.
Topic 5
Industry applications and case studies
| Industry | Example |
|---|---|
| Finance | Banks analyse transaction patterns to flag fraud in real time |
| Marketing | E-commerce sites analyse browsing to recommend products and target ads |
| Operations | Retailers forecast demand to keep the right stock and cut waste |
Example
A food delivery company studies order data and finds that rainy evenings double demand; it adds delivery partners on rainy days — a direct operations decision from data.
Key terms
- Data cleaning
- Fixing or removing incorrect, incomplete or duplicate data
- Missing value
- A blank where data is expected
- Outlier
- A value far away from the rest of the data
- IQR
- Interquartile range, Q3 − Q1
- Data ethics
- Responsible, fair and lawful use of data
Quick revision
- Formats: CSV, XLSX, JSON, XML.
- Clean: duplicates, formats, errors, missing values, outliers.
- Outlier rule: outside Q1 − 1.5 IQR to Q3 + 1.5 IQR.
- Ethics: consent, privacy, security, fairness, transparency.
Important exam questions
Practice questions written to the PTU exam pattern for this unit's syllabus: short answers (Section A style) and long answers (Sections B and C style).
Short-answer questions
- Q1.Name three common data formats.
- Q2.What is data cleaning?
- Q3.List two ways to handle missing data.
- Q4.What is an outlier?
- Q5.State two ethical issues in data analytics.
Long-answer questions
- Q1.Explain the steps of data cleaning and transformation with examples.
- Q2.Explain methods of handling missing data and outliers.
- Q3.Discuss ethical considerations in data analytics.
- Q4.Explain applications of data analytics in finance, marketing and operations with case studies.
Stuck on this unit?
Message SBS on WhatsApp for help with Basics of Data Analytics using Spreadsheet, or to ask about studying BCA at Synetic.
