Unit 5 · Data · Lesson 5.13 · Review

Rebuild the Map & Review

Unit 5 closes out Big Idea 2. Put it on the map, drill the claim relay until “supported / not supported / need more” is automatic, and take a System Check across everything so far.

Big Idea 2 — DataReview — Units 0–5

Before this lesson — read on Runestone

18.9 Chapter 18 — Concept Summary

Two minutes. A term that feels fuzzy there is tonight’s study target. Enrolled in our Runestone course? Open it from there so your progress counts.

Where We Are

People→ Bits→ Programs→ Internet→ Security→ Big Data→ Impact

Big Data is on the map now, between Bits and Impact: Unit 1 taught how bits represent things; Unit 5 taught what you can learn from piles of them — and what you can’t. That second half leads straight to Unit 8.

The Point of Today

Can you say the unit in one breath? “A dataset is records and fields someone chose; you clean it, filter and group it, and chart it honestly — and then you say only what it supports, because correlation isn’t causation and the data never includes everyone.” If one piece is fuzzy, you just found tonight’s study target.

Retrieval, not re-reading. Produce the answer, then check.

The Unit 5 Map — four regions

Cover the right column and define each term before you peek.

Region	Terms
What data is (5.1–5.3)	data, dataset, record, field, structured data, metadata, CSV, delimiter, header row, parsing
Getting answers (5.4–5.7)	data cleaning, missing data, outlier, filter, aggregate
Showing answers (5.8)	visualization, axis, scale — accurate vs honest
Limits (5.9–5.10)	correlation, causation, confounding variable, bias, sample, representative

The unit’s spine is one three-way judgment. Every claim from data is one of: S — supported as stated · X — not supported (usually cause claimed from correlation, or a biased sample treated as everyone) · ? — need more information, and you can name what.

Claim-Evaluation Relay

The most valuable review activity of the unit. File each claim. Wrong files explain themselves.

Vocabulary Rapid-Fire

The eight terms most likely to show up — and the three students blur together are first. Say the definition out loud, then check.

metadata Big Data: Data about data — date, size, author, location. Often more revealing than the content.
confounding variable Big Data: A third factor that drives both variables, creating a correlation without causation.
representative Big Data: A sample that matches the whole population on the traits that matter.
data cleaning Big Data: Handling missing, placeholder, inconsistent, and duplicate values — with the rules written down.
filter Big Data: Keep only the records that meet a condition.
aggregate Big Data: Combine many values into one: count, sum, average, max, min.
correlation Big Data: Two variables tend to change together. Not proof of cause.
bias Big Data: A systematic tilt in data or results, often from who was left out.

System Check

Ten questions from Units 0–5, weighted to Big Idea 2 since this unit closes it out. Trace on paper.

0 of 10 answered

Connections

Pulls together 5.1–5.12 — a claim you misfiled above names the lesson to reopen: 5.9 for X/?, 5.10 for biased samples.

Next in 5.14 — the Unit 5 test. Tomorrow’s page has the breakdown and a practice set.

Closes Big Idea 2 — bits (Unit 1) and data (Unit 5) are now one region of the map.

Opens Unit 6 — all this data has to travel somewhere. Next: the internet.