Year 9 Science · Unit 4 · Lesson 11

Using a Large Dataset to Develop and Test a Question

Challenge Worksheet

Name

Date

Class

Learning Goals

Critically evaluate whether a question can be answered by a given dataset.
Justify cleaning decisions using the ideas of outliers, missing values and duplicates.
Design and defend a full workflow from a raw dataset to an evidence-based answer.

Warm Up

Odd one out

Circle the item that does not belong in each group. Then explain why it doesn't fit in the answer column.

#	Group	Your answer (odd one + reason)
1	Explore the columns Form a question Clean the data Guess the answer first
2	Missing value Duplicate row Impossible outlier A surprising but real value
3	Bureau of Meteorology Australian Bureau of Statistics data.gov.au Data you measured yourself
4	Row is a record Column is a variable Header names the units Every row is the answer

Go Deeper

Evaluate the plan

A student proposes...

"I downloaded a 12,000-row Bureau of Meteorology file with columns for date, maximum temperature and rainfall. I want to find out whether hot weather makes people buy more sunscreen, so I will scroll to a few rows that look hot and write my conclusion from those. I noticed one temperature reading of 480 degrees, so I will keep it because all data must be kept exactly as found."

(a) Explain why this dataset cannot answer the student's question as written, and what change to the question or data would be needed.

Challenge 3 marks

(b) The student plans to "scroll to a few hot rows" and keep the 480-degree reading. Identify two separate things wrong with this approach, and explain the correct action for each.

Challenge 3 marks

(c) Rewrite the whole task as a sound investigation using a dataset of your choice. State the columns it would need, an investigable question, the independent and dependent variables, one cleaning step, and how you would reach an evidence-based answer.

Challenge 4 marks

Wrap Up

In one sentence, what was the main idea of this lesson?