Skip to content
FORGENO ADS GAMESUnlock all $8

Data Lab · Unit 1 of 8

Unit 1: Thinking with data

What data is, how tables are organized, the kinds of values they hold and how to turn a hunch into a question data can answer.

Opens the first lesson of this unit in Data Lab. Every lesson is open from the start.

Lesson 1: What data is, and why it matters

Goal: Explain what data analysis is and what an analyst actually does.

Data is recorded facts: sales, temperatures, test scores, clicks, survey answers, sports stats. On its own, a pile of data says very little. Data analysis is the work of turning it into answers people can act on.

Almost every job now involves some data: a store manager deciding what to restock, a nurse tracking patient readings, a coach comparing players, a YouTuber reading which videos keep viewers watching. Being the person who can find the answer in the numbers is a valuable skill.

The work usually follows the same cycle:

  1. Ask a clear question.
  2. Get the data that can answer it.
  3. Clean it so it is accurate and consistent.
  4. Analyze it: calculate, group, compare.
  5. Show it with tables and charts.
  6. Decide: say what it means and what to do.

Worked example: a lemonade stand

  1. Ask: which day of the week should we open longer?
  2. Get: cups sold each day for four weeks.
  3. Clean: fix one day typed as "4O" (letter O) instead of 40.
  4. Analyze: average cups for each day of the week.
  5. Show: a bar chart of the averages.
  6. Decide: Saturdays average 52 cups and weekdays about 20, so open longer on Saturdays.

Even a tiny business follows the same cycle a big company does.

Data
Recorded facts or measurements, such as numbers, text, dates or categories.
Data analysis
Turning raw data into answers by asking questions, cleaning, calculating, visualizing and interpreting.

Remember: Ask, get, clean, analyze, show, decide. Most data mistakes come from skipping the first or third step.

Lesson 2: Rows, columns and tidy tables

Goal: Read a data table and set one up the tidy way.

Most data lives in tables. In a well-organized table, called a tidy table:

  • Each row is one record: one sale, one student, one day.
  • Each column is one field: one kind of information, like date, item or price.
  • Each cell holds one value.
  • The first row holds clear column names (headers).
A tidy table: one row per order, one column per field
order_iddateitemqtyprice
10012025-03-01Notebook33.50
10022025-03-01Backpack135.00
10032025-03-02Notebook23.50

Messy tables cause most of the pain in data work. Common problems: two facts squeezed into one cell ("3 notebooks"), a separate column for every month, totals mixed in with the data rows, merged cells and colors used to mean something.

Worked example: fixing a messy column

  1. A cell says 3 notebooks.
  2. That one cell holds two facts: a quantity and an item.
  3. Split it into a qty column (3) and an item column (Notebook).

Now you can add up quantities and count items, which you could not do while they were stuck together.

Row (record)
One complete entry in a table, such as one order or one person.
Column (field)
One kind of information recorded for every row, such as date or price.
Tidy data
A table where each row is one record, each column one field and each cell one value.

Remember: One row per record, one column per field, one value per cell.

Lesson 3: Kinds of data

Goal: Tell numbers, categories, text and dates apart, and know why it matters.

Every column holds a certain type of data, and the type decides what you can do with it.

TypeExamplesWhat you can do
Numbers (numeric)price, age, temperature, quantityAdd, average, find the biggest and smallest
Categoriescity, size (S/M/L), yes/no, teamCount how many in each group, compare groups
Textcomments, names, descriptionsSearch, count words, read
Dates and times2025-03-01, 14:30Sort in time order, group by month, find time between

A common trap: some numbers are really labels. A ZIP code, a phone number or a student ID is made of digits, but adding or averaging them is meaningless. Treat them as categories or text.

Another trap: numbers stored as text. If a spreadsheet shows a number lined up on the left, or a total ignores some values, it may be reading those cells as text, for example because of a stray space or a "$" typed in.

Dates have a trap too. "03/04/2025" means March 4 in the United States and 3 April in much of the world. The format 2025-03-04 (year-month-day) is unambiguous and sorts correctly, which is why analysts like it.

Numeric data
Values you can do arithmetic on, such as prices or quantities.
Categorical data
Values that put records into groups, such as city or size.
ISO date format
Writing dates as YYYY-MM-DD, which is unambiguous and sorts in time order.

Remember: Ask "does arithmetic make sense on this?" If not, treat it as a category, even if it is made of digits.

Lesson 4: Asking questions data can answer

Goal: Turn a vague goal into a specific, measurable question.

The quality of an analysis depends on the quality of the question. Vague questions lead to vague answers.

VagueSpecific and measurable
"Are our customers happy?""What percentage of customers rated us 4 or 5 stars in the last 3 months?"
"Which products are good?""Which 5 products brought in the most revenue last year?"
"Is the new menu working?""Did average weekly sales change in the 8 weeks after the new menu, compared with the 8 weeks before?"

A good analysis question usually names:

  • What you will measure (a number you can calculate)
  • Who or what is included (all customers? only members?)
  • When (which time period)
  • Compared with what (last year, another group, a goal)

Then list the columns you would need. If you cannot name the columns, the question is still too vague, or you need to collect new data.

Worked example: a school club

Goal: "Get more people to come to robotics club."

  1. Question: "Which month had the highest average attendance this year, and what was different about it?"
  2. Data needed: date of each meeting, number attending, any event or announcement that week.
  3. This points to a real action: repeat whatever happened in the best month.

Remember: A good question says what you will measure, for whom, when, and compared with what.

Practise it in Data Lab

Each lesson ends with two or three questions that check you can use the idea, not just recognise it. Wrong answers explain why and come back in Review. The unit ends with a 5-question quiz; pass it with 80% or more to master the unit.

Every lesson is free: no ads, no lives and no paywall.