Data Tables (Tabular Data)#

📦 Data Preparation 🧬 Data Types & Structure Lesson 006

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

The shape analysis lives in#

Most data analysis happens on tabular data — data arranged in a table of rows and columns. It is the shape spreadsheets display, databases store, and the analysis tools of this course expect. Understanding what makes a table well-formed is the bridge between raw data and analysis, and it formalises the layout rules the spreadsheet-organisation lesson introduced.

The anatomy of a data table#

A proper data table has a consistent structure:

Rows are records (also called observations) — each row is one instance of the thing being measured: one sale, one patient, one day.
Columns are variables (also called fields or attributes) — each column is one measured property, holding the same type of value down its length: date, region, amount.
A header row names the columns, giving each variable an explicit, unique label.
Cells hold one value each — the intersection of one record and one variable.

The defining discipline is consistency: every row has the same columns, and every column holds one kind of value. That regularity is exactly what lets a tool sort, filter, aggregate, or join the table mechanically.

Why the structure is non-negotiable#

A table that breaks the pattern breaks the tools. Two variables crammed into one column (“New York, NY” where city and state should be separate) cannot be grouped by state. One record split across two rows double-counts under aggregation. A column mixing types defeats sorting and math. The rules are not aesthetic — they are the contract every downstream operation relies on, which is why cleaning so often means restoring proper tabular structure before anything else.

Tables and the relational world#

The tabular shape is also the foundation of relational databases, where data lives in many linked tables — a topic the sources stage of this section develops. The same “rows are records, columns are variables” discipline scales from a single spreadsheet to a database of hundreds of related tables; learning it here is learning the grammar of both.

The caveat#

Not every table you receive is well-formed — spreadsheets in the wild are full of merged cells, sub-headers, totals rows wedged between data, and summary layouts meant for human eyes rather than analysis. Recognising the gap between a presentation table and an analysis table is a core preparation skill: the next lesson examines two legitimate analysis layouts — wide and long — and when each is the right one.

Hint