Data Analytics Practice in Kalyan: Build an Attendance Report You Can Trust

Practise data cleaning and reporting with fictional attendance records. Separate absent, cancelled and unknown entries and explain your denominator.

An attendance report looks simple until you ask what each number means. An absence, a cancelled class and a missing attendance entry are different situations. Combining them can produce a misleading result. For someone exploring Data Analytics or Data Analyst training in Kalyan, this exercise offers a practical way to discuss data preparation, validation and reporting.

Start with fictional records

Use invented learner codes, sessions and observations. Do not use real names, contact information or institute records. This is a suggested learning exercise, not a claim about an IT Accurate class or student project.

The expected-attendance table matters. Without it, you might incorrectly count every session for every learner, including sessions before a learner joined.

Define the question

Ask which fictional learners may need follow-up after missing completed sessions. Exclude cancelled sessions from the expected count. Keep missing observations unresolved until checked; do not silently treat them as absent or present. Define late-arrival and partial-attendance rules consistently.

Work through one example

Learner L001 has four scheduled sessions. One is cancelled. Of the three completed sessions, one has a present observation, one an absent observation and one no observation. The report should show:
Expected completed sessions: 3.
Present: 1. Absent: 1. Unresolved: 1.

Present divided by expected completed sessions is 1/3. Present divided by known observations is 1/2. These answer different questions. Never display either percentage without explaining the denominator and the unresolved observation. Raw counts are often clearer while an entry remains missing.

Check before making a dashboard

Keep the original records and a cleaning log. A repeated observation may be an import error or a correction; resolve it explicitly rather than deleting it without an audit trail.

Use these seven acceptance checks:
1. A cancelled session adds nothing to expected completed sessions.
2. A missing observation adds to unresolved, not absent.
3. Duplicate observations do not double-count a session.
4. A learner joining later does not inherit earlier expected sessions.
5. Corrections update the reporting value while preserving the original record.
6. No expected completed sessions means not applicable, not 0% attendance.
7. Every total traces back to the expected-session list and observations.

Make uncertainty visible

Show the reporting period, expected completed sessions, present, absent and unresolved counts together. Do not compare someone expected to attend twice with someone expected twenty times without explaining that difference. A report should help a reviewer identify the next check, not disguise incomplete data.

Explain your work

Prepare a five-minute explanation covering the business question, table relationships, one data-quality issue, a manually checked result and a limitation. Include a data dictionary and cleaning log. The exercise should demonstrate reasoning another person can inspect.

Discuss the learning path

If you are exploring Data Analytics training in Kalyan, ask IT Accurate how the current syllabus covers data cleaning, SQL, reporting and feedback on practice. Confirm the tools, learning mode and batch timings with the branch. Contact us for fees.