Sourcing Data

Gathering Data

Students need to know about different types of data, as well what a sample and a population are.

Tasks/Ideas

  • Leckie Higher Applications Student Book – Chapter 6 – Exercise 6A
  • Types of Data Worksheet- Download Here
  • Populations and Samples- Mrs Cao’s Classroom
  • A great activity is to have students search online for a dataset that interests them. Answer the following:
    1. Have them look at each column- what are the data types?
    2. Is the dataset a sample, if so what was the population?
    3. If this was a sample, was it a random sample? How do you know?

Bias

Tasks/Ideas

  • Leckie Higher Applications Student Book – Chapter 6 – Exercise 6B
  • This worksheet has lots of different scenarios to consider, with answers too!

Outliers

Tasks/Ideas

  • Leckie Higher Applications Student Book – Chapter 6 – Exercise 6C
  • How does removing outliers affect different statistics? – Outliers – Applying Maths

Sourcing Data for the Project

Tasks/Ideas

  • Leckie Higher Applications Student Book – Chapter 6 – Project Box
  • Here are some useful places for students to look for data:
    • Open Intro – Lots of data on lots of topics, all available in a CSV format. Most importantly for students doing their projects, it describes where the data comes from. I couldn’t recommend this website more!
    • CORGI Data Sets – The Collection of Really Great, Interesting, Situated Datasets. Nothing to do with dogs, just lots of students-friendly datasets to use and explore.
  • A good piece of advice is to search for datasets using the terms “csv” and “beginner” in the search terms. Another piece of advice- if the dataset looks like a mess, it’s not clear whether it’s a sample or a population, and you can’t find any information about where the dataset came from- leave it be and keep searching!

Remember, it makes no sense to complete a hypothesis test on a population. You can only complete a hypothesis test on a sample. As the project requires a hypothesis test to be completed, the dataset must be a sample. Never lose sight of this!