Statistics - lesson 3 | HackIndia Skills

Key point

The sample is the claim

Your insight is only about the people, events, rows, or products your sample can fairly represent.

If you collected responses from 42 students in your own WhatsApp groups, do not write “Indian students prefer X”. Write “In our sample of 42 college students, mostly from our network, X was common”.

Population vs sample

The population is the full group you want to talk about. The sample is the part you actually observed.

For a hackathon project, your population might be “students using UPI for canteen payments”. Your sample might be “78 students from two colleges who answered our form”.

Your to-do

Check sample fit

Do each one yourself, then tap it to tick it off. The ticks are only a checklist for you: they are not marked or scored.

Name the population you want to describe.
List where each row or response came from.
Compare your sample to the population on obvious traits like college, year, city, device, language, or time of day.
Mark who is missing or less likely to answer.
Rewrite the insight so it only claims what the sample supports.

0 of 5 done

Common mistake

Common overclaim

Beginners often treat easy data as if it represents everyone. This is convenience sampling. It is fine for a hackathon if you label it clearly.

Do not hide it. Recruiters trust a builder who says “our sample is biased toward engineering students” more than one who makes a big claim from a small form.

Example

Sample note example

Add a small note beside your chart or in your README. This turns a weak claim into an honest one.

Here is a simple project note for a canteen payments feature.

Dataset / sample
Source: Google Form shared in 3 college WhatsApp groups.
Rows: 78 responses collected on 12 Jan 2026.
Who is included: Mostly engineering students from Pune and Bengaluru.
Who may be missing: Non-engineering students, cash-only users, students not active on WhatsApp.
Use this for: Early product direction and demo decisions.
Do not use this for: Claiming behaviour of all Indian college students.
Insight: In this sample, UPI was the most common canteen payment method.
Limitation: The sample is small and biased toward our own network.

Tip

Use AI carefully

You can ask ChatGPT, Gemini, or Claude to inspect your sample description before submission.

Paste only non-sensitive summary details. Check that the AI does not invent representativeness, percentages, or a population you did not measure.

Prompt:
I am submitting a hackathon project with this sample note. Tell me if my insight overclaims beyond the sample. Suggest a safer wording.

Sample note:
[Paste your note]

Insight:
[Paste your insight]

What good looks like

A good sample section is short. It says source, size, time, included groups, missing groups, and a safe claim.

You do not need a perfect random sample for a hackathon. You need a clear boundary around what your data can and cannot say.