แบบกว้างกับแบบยาว และข้อมูลเป็นระเบียบ

The same numbers can sit in very different shapes. Mastering data layout is the first step to reshaping like a pro. 📐

In wide format, each category gets its own column, so one row holds many related values side by side.

In long format, those columns collapse into rows. One column names the variable, another holds its value.

Nothing is lost going wide to long or back. You are only rearranging where each value lives, not changing any numbers.

Tidy data is a tight version of long format: each variable is a column, each observation a row, each value one cell.

Most pandas tools expect tidy input. When data is tidy, grouping, filtering, and plotting just work without awkward reshaping.

If column names are really values, like dates or product names, the table is untidy and probably needs to go long.

Wide is great for humans to read and for cross-tab summaries. You often analyze in long, then pivot to wide for a final report.

Pandas gives you a small kit: melt goes wide to long, while pivot and pivot_table go long back to wide.

Choosing a layout is a design decision. The right shape can turn a tangled analysis into one or two clean lines.

Ask one question of any table: does each row mean exactly one observation? If yes, you are probably already tidy.

Time to test your layout instincts.

You learned wide versus long and why tidy data powers pandas. Next you will pivot long data into clean cross-tabs. 🎯