Skip to content
PlayOwl

Computers & Digital Literacy · Grade 9

Representation, Sampling and Bias

Learners turn categories and text into numbers with one-hot encoding and word counts, and rescale numeric features with min-max scaling using training-data limits. They then examine how the data a model learns from can mislead it: sampling variability versus representation bias (with a sampling simulation), and measurement, label and historical bias in fictional datasets, each paired with a mitigation.

  • 4 lessons
  • About 105 minutes
  • Representation and bias check: 9 questions
  • Ways to answer: fill in the blanks, choose an answer, trace code, estimate, highlight a text and match pairs
Try it now (no account) One of the five free starter courses Open in PlayOwl

What learners will discover

  • Encode categorical features with one-hot encoding and text with word counts, and explain why arbitrary integer codes can mislead a model.
  • Apply min-max scaling, (x − min) ÷ (max − min), using the training data's minimum and maximum, and explain why scaling matters for distance-based models.
  • Distinguish sampling variability from representation bias, and compare a dataset's makeup with the population a model will serve.
  • Identify measurement, label and historical bias in fictional cases and propose a matching mitigation.

Lessons

  1. Turning data into numbers 7 steps5 practiceabout 26 min
  2. Scaling numeric features 7 steps5 practiceabout 24 min
  3. Samples, variability and representation bias 7 steps4 practiceabout 28 min
  4. Measurement, label and historical bias 6 steps3 practiceabout 27 min
  5. Representation and bias check 9 questions at the end

Keep exploring

More 9th grade computer skills

Guide for grown-ups