Chapter 5: Understanding Data
Original notes — ready CBSE (NCERT, English Medium) • Class 11 • Computer Science • Book: Informatics Practices
Chapter 5: Understanding Data — EG's own notes, in simple words
Book: Informatics Practices (CBSE, Class 11) | Chapter 5 | Medium: English
Diagram (labels on the image): Diagram (labels on the image): understanding data — raw messy data, a cleaning step, and a clean table ready for analysis.
Check before you compute: clean data is what makes an answer trustworthy.
1. 1. Data, and why it needs understanding
Data is only useful when it is understood: its source, its type, its quality and its limits. A dataset of student marks is useless if nobody knows which exam it came from, whether marks are out of 100, or whether absent students are marked zero or left blank. Understanding data means asking those questions before computing anything.
Data can be structured (tables with rows and columns), semi-structured (labelled text such as web data) or unstructured (free text, photos, audio). School and office work mostly uses structured data, which is why tables dominate this book.
2. 2. Types of data
Data types describe what values mean. Numeric data splits into integers (counts) and floats (measurements). Text data holds names and labels. Boolean data holds True/False. Dates and times are their own kind, because computing with them needs calendar rules.
Categorical data names groups (house colour, subject, grade), while numerical data measures or counts. Confusing the two leads to nonsense averages: averaging roll numbers is meaningless, because roll numbers are labels, not quantities.
3. 3. Data quality: the hidden exam topic
Good data is accurate (correct values), complete (no missing entries that matter), consistent (same format everywhere) and timely (fresh enough for the decision). Common problems are missing values, duplicate records, wrong formats (dates written two ways) and outliers (a mark of 1000 out of 100, likely a typing error).
Cleaning means fixing these problems before analysis: filling or flagging missing values, removing duplicates, standardising formats, and checking outliers instead of silently deleting them.
4. 4. From data to decision
Analysis turns clean data into information: totals, averages, comparisons and simple charts. A bar chart of average marks by subject shows at a glance which subject needs attention; a table of attendance by month shows trends. The final step is always a decision: extra classes, a timetable change, or a message to parents — data that leads nowhere is unfinished work.
At a glance — chart / table
| Quality problem | Example | Cleaning action |
|---|---|---|
| Missing value | Blank mark for a present student | Fill from record or flag as missing |
| Duplicate record | Same student entered twice | Remove the duplicate |
| Wrong format | Dates as 5/6 and June 5 mixed | Standardise to one format |
| Outlier | Mark of 1000 out of 100 | Check source; correct if a typo |
Key Terms / Important Terms
- Structured data — Data organised in rows and columns, like a table.
- Categorical data — Data naming groups or labels, e.g. subject or grade.
- Numerical data — Data measuring or counting quantities.
- Missing value — An absent entry in a dataset.
- Outlier — A value far from the rest, possibly an error.
- Data cleaning — Fixing quality problems before analysis.
- Data analysis — Turning clean data into information for decisions.
Remember
- Ask source, type, quality and limits before computing anything.
- Roll numbers and grades are labels — never average them as quantities.
- Quality = accurate, complete, consistent, timely.
- Clean first, analyse second: missing values, duplicates, formats, outliers.
Exam Tip / Common Mistake
Common mistake: averaging categorical labels such as roll numbers or grade codes. If the number is a label rather than a quantity, an average of it means nothing — say so in the answer.
Practice set — try it yourself
- A dataset mixes dates written as 5/6 and June 5. Name the quality problem and the fix. Hint: Hint: Same information in different formats.
- Why is averaging roll numbers meaningless? Hint: Hint: Roll numbers label students; they are not quantities.
- Give one example each of structured and unstructured data in a school. Hint: Hint: A marks table vs a photo of a function.
- A mark of 1000 out of 100 appears. What should an analyst do first? Hint: Hint: Check the source before deleting or keeping it.
- Name four data-quality checks and one example of each failing. Hint: Hint: Accuracy, completeness, consistency and timeliness.
Questions with Answers
- What is structured data? Give a school example. Answer: Answer: Structured data is organised in rows and columns like a table, for example a marks table with one row per student and one column per subject.
- Distinguish categorical from numerical data with examples. Answer: Answer: Categorical data names groups, such as subject (Hindi, Maths) or grade (A, B). Numerical data measures or counts, such as marks (88) or attendance days (24).
- What is data cleaning, and why does it come before analysis? Answer: Answer: Data cleaning is fixing quality problems — missing values, duplicates, wrong formats and outliers — before analysis. It comes first because analysis of dirty data gives misleading totals, averages and decisions.
- A dataset has the same student entered twice. What problem is this, and what is the fix? Answer: Answer: This is a duplicate record, a completeness/consistency problem. The fix is to remove the duplicate entry so the student is counted once.
- How does analysis turn data into a decision? Give an example. Answer: Answer: Clean data is summarised into information, such as average marks by subject, which is read from a table or chart. The decision follows from it, such as arranging extra classes for the weakest subject.