University of Rajasthan Students: Writing a Data Analytics Internship Viva Answer Bank

To pass a data analytics internship viva, you need to explain your own project in plain words: the question you asked, the data you used, the cleaning and analysis choices you made, and what the result does and does not prove. This article gives you 30 likely questions grouped into project, tools and ethics, an outline for each answer, a 14-day preparation plan and a rubric for scoring each other.

What a data analytics internship viva actually tests

A viva is a spoken examination. The examiner may be a faculty member, a department coordinator or a mentor, and the questions rarely come from a textbook. They come from your report. The examiner wants to know whether you did the work, whether you understand it, and whether you can communicate it to a non-technical person.

In data analytics, three things get probed again and again:

• Ownership: Can you describe each step and say why you chose it?
• Judgement: Do you know where your data was weak, and what you did about it?
• Honesty: Can you say what your analysis cannot claim?

Short, specific answers beat long, rehearsed ones. A good answer is usually three to six sentences, with one concrete detail from your own work.

A 14-day plan that builds viva material as you work

A two-week internship is short, so viva preparation has to run alongside the project. The free two-week online internship from CareerSwami comes with a free offer letter and a free completion certificate after successful completion, and its 14-day length suits this plan. Whichever programme you do, keep notes daily; you cannot rebuild them from memory on day 15.

Project questions: 12 likely viva questions with answer outlines

These test whether the project is really yours. Replace the outlines with details from your own work.

1. Summarise your project in one minute. Outline: problem, data, method, main finding, one limitation.
2. What problem were you trying to solve, and who would use the answer? Outline: name a realistic user (a shop owner, a college office) and the decision they would make.
3. Where did your data come from? Outline: source, collection date, size, whether it is public, and its licence or terms of use.
4. How many rows and columns did you have, and what did each represent? Outline: give the real numbers and the unit of observation.
5. How did you handle missing values? Outline: how many were missing, in which columns, the method (drop, fill with median, flag) and why it suits that column.
6. Did you find outliers? What did you do? Outline: how you detected them (box plot, IQR), whether they were errors or real, and what changed after treating them.
7. Which insight surprised you most? Outline: the finding, the evidence, and whether you checked it was not a data error.
8. Why did you choose this chart or metric? Outline: match the chart to the question (trend over time suggests a line chart) and say what you rejected.
9. How do you know your result is reliable? Outline: sample size, cross-checks, a sanity check against a known figure, a train/test split if you modelled.
10. What are the limitations of your analysis? Outline: data coverage, time period, missing variables, and what you would need to strengthen the claim.
11. What would you do with another month? Outline: two specific steps, such as more data or testing a second method.
12. What was your own contribution if you worked with others or used a template? Outline: state exactly which parts you wrote and which were provided.

Tools questions: 10 likely viva questions on Excel, SQL, Python and BI

Answer only about tools you actually used. Saying “I used Excel for cleaning and a pivot table for the summary” is stronger than listing six tools you touched once.

13. Why did you choose this tool for this task? Outline: fit to data size and task, your existing skill, and what the alternative would have cost.
14. What is the difference between a pivot table and a formula-based summary? Outline: pivots aggregate and regroup quickly; formulas give fixed, traceable logic.
15. Explain VLOOKUP or XLOOKUP and one way it can fail. Outline: lookup by key; fails on duplicate keys, extra spaces or mismatched data types.
16. What is the difference between INNER JOIN and LEFT JOIN? Outline: inner keeps matches only; left keeps every left row and fills gaps with NULL. Give an example from your tables.
17. What do GROUP BY and HAVING do, and how do they differ from WHERE? Outline: WHERE filters rows before grouping; HAVING filters groups after aggregation.
18. In pandas, how would you find and remove duplicates? Outline: duplicated() to inspect, drop_duplicates() to remove, and checking which key defines a duplicate.
19. What is the difference between mean, median and mode, and when did you use each? Outline: median for skewed data such as income; mean for roughly symmetric data; mode for categories.
20. What does correlation tell you, and what does it not? Outline: strength and direction of a linear relationship; not causation; may be driven by a third variable.
21. How did you design your dashboard? Outline: the audience, the two or three key metrics at the top, filters, and removal of clutter.
22. How did you make your work reproducible? Outline: saved scripts or notebooks, documented steps, original data kept untouched, file naming.

Ethics and data responsibility: 8 likely viva questions

Examiners increasingly ask these because they separate careful analysts from careless ones. Keep answers practical.

23. Did your dataset contain personal information? Outline: say yes or no; if yes, explain what you removed or masked (names, phone numbers, addresses).
24. Do you have the right to use this data? Outline: public or open licence, permission, or terms of the platform; never data taken from a private source without consent.
25. Could your analysis be biased? Outline: who is over- or under-represented, and how that could skew the conclusion.
26. How did you avoid cherry-picking results? Outline: you reported findings that did not support your expectation, and kept the full analysis log.
27. How would you explain uncertainty to a non-technical reader? Outline: use ranges and plain words (“likely”, “based on this sample”) instead of absolute claims.
28. Could a chart you made mislead someone? Outline: truncated axes, selective time windows and dual axes are common traps; say what you did to avoid them.
29. What would you do if a mentor asked you to adjust results to look better? Outline: explain the finding, show the evidence, decline to alter data, and document the discussion.
30. How did you protect and store the data? Outline: no sharing on public links, no uploading sensitive files to unknown tools, deletion after use where appropriate.

Worked example: shaping one project answer from a rough draft

Illustrative only; the numbers are invented to show structure, not real results.

Rough answer: “I analysed sales data and made a dashboard. It shows sales are going up.”

An examiner will follow with “Which sales, over what period, and how do you know?” and the answer collapses. Rebuild it using the problem, data, method, finding and limitation pattern:

Improved answer: “I looked at monthly sales for a hypothetical stationery shop to see which product categories drive revenue. The dataset had 12 months of invoices. I removed duplicate invoice numbers and filled missing category labels as ‘Unknown’ instead of deleting them, so totals stayed intact. A pivot table showed one category producing the largest share of revenue, and a line chart showed a rise in June and July. The limitation is that one year cannot separate a seasonal pattern from a real trend, so I would want a second year before recommending stock changes.”

The improved version names the data, shows a cleaning decision with a reason, states a finding, and admits a limit. Each sentence is something you can be asked about and answer.

Self-scoring rubric for practising the viva with a friend

Ask your friend to pick questions at random from the bank, interrupt with one follow-up, and score each answer on the five criteria below. Score from 0 to 4 per criterion, giving a maximum of 20 per answer.

Reading your practice scores

• 16-20: ready; keep the answer short.
• 11-15: add one concrete number or example to the answer.
• 6-10: revisit the underlying step in your notebook before practising again.
• 0-5: relearn the concept, then rewrite the answer in your own words.

Practise at least 15 questions across the two final days, covering all three groups. Track which criterion you lose points on most; for many students it is the link to their own project.

University of Rajasthan students: internship rules and viva preparation

The University of Rajasthan describes itself as having begun as an affiliating university with twenty-two colleges and as now holding “University with Potential for Excellence” status, according to its official history page (an excerpt; check the live page). It also runs constituent colleges such as University Maharaja College, University Commerce College and University Law College. In practice, this means your internship expectations may depend on whether you study in a university department or an affiliated college.

A university notice dated 27 June 2023 mentions an internship duration of 120 hours or 3 weeks for 4 credits, and refers to a summer internship in the first year. This comes from a search excerpt, so the programme it covers, the evaluation method, the documents needed and whether online internships qualify are all unconfirmed. A 14-day internship may not match a 120-hour or 3-week requirement by itself, so do not assume it will count for credit.

For the viva itself, the university’s syllabus page lists PG syllabi for 2025-27, and the University Centre for Computer Science & Information Technology publishes an MCA syllabus PDF for semesters I-IV in its 2026 folder. Use these to align your tool and concept answers with what your course teaches. The Central Placement Cell page describes training sessions on personality development and career counselling; it looks dated, so confirm whether sessions are running now.

Verification steps for UNIRAJ students before your viva

1. Ask your department office or internship coordinator in writing whether the 27.06.2023 notice applies to your programme and batch.
2. Ask whether an online internship of 14 days is accepted, and whether it counts for credit or is only extra experience.
3. Request the report format, viva panel, marking scheme and submission deadline as a written document.
4. Ask who evaluates the viva (internal examiner, external examiner or mentor) so you can pitch your answers appropriately.
5. Check your own syllabus PDF for the data, statistics or programming topics most likely to be asked.
6. Email the placement cell (details on its page) to ask whether current training sessions or mock vivas are available.

Final-week habits that raise viva scores

• Keep a one-page sheet with your dataset size, key numbers and three findings so you never quote a figure wrongly.
• Rehearse aloud, not silently; spoken answers expose gaps that reading hides.
• Prepare a 60-second and a 3-minute version of your project summary.
• Open your notebook or dashboard before the viva, in case you are asked to show a step.
• When you do not know something, say what you do know and how you would find out. This is better than guessing.

Frequently Asked Questions

How long should each viva answer be?

Aim for 30 to 60 seconds, which is roughly three to six sentences. Give the direct answer first, then one detail from your project. Examiners will ask a follow-up if they want more.

What if the examiner asks about a tool I did not use?

Say plainly that you did not use it in this project, then explain what you did use and how the tasks compare. Do not bluff, because follow-up questions will expose it.

Do I need to memorise code for the viva as a University of Rajasthan student?

No. You should understand the logic of each step and be able to describe the function or query in words. Memorising exact syntax matters less than explaining why you used it.

Which questions are asked most often?

There is no verified list for any specific panel. Questions about your project summary, data cleaning, chart choices and limitations are the safest to prepare thoroughly, since they follow directly from any report.

Can a 14-day online internship count for credit at the University of Rajasthan?

That is unconfirmed. A university notice mentions 120 hours or 3 weeks for 4 credits, but its scope is unclear. Ask your department in writing before assuming any credit.

What does CareerSwami’s internship provide?

It is a free two-week (14-day) online internship with a free offer letter and a free completion certificate after successful completion, across 26 tracks. It does not promise a stipend, academic credit, placement or jobs.

How should I practise if I have no friend available?

Record yourself answering random questions on your phone, play it back and score it with the same rubric. Pay attention to filler words, vague phrases and answers that lack a project detail.

What should I do if my project result was weak or inconclusive?

Say so. Explain what the data could and could not show, what you checked and what you would try next. A candid account of an inconclusive result often scores higher than an overstated one.

Related Data Analytics guides

• Arunachal University of Studies: Learning Power BI in Two Weeks Before Your Internship
• Sai University Students: Picking Between a Startup and a Corporate Data Analytics Internship
• UKA Tarsadia University Students: Documenting Data Sources and Assumptions in Analytics Work
• All Data Analytics internship guides