Statistical Reasoning

Null Hypothesis Testisn

Klinkenberg

University of Amsterdam

2024-02-05

Course structure

Course parts

  • Lectures: On campus / Online / Video recording
  • Preparatory Assignment: Submit in Testvision and Reflection in canvas
  • Tutorials: For your support and in class assessment
  • Exam: Knowledge and skills

in class assessment

Grading

Final grade=0.9×exam grade+0.1×credit points

  • Exam
  • Credits for
    • Tutorial participation
    • Preparatory assignment

Credits

You get 10 at the start of the course.

Number of times missed Deduction for TU Deduction for PA
1x 0 0
2x 0 2
3x 2 3
4x 4 4
5x or more 5 5

See canvas for all the details.

Learning

Reasoning in statistics

Source: ARTIST

Statistical Literacy

  • Knowledge (Basic understanding of concepts)
    • Identify
    • Describe
  • Skils (Ability to work with statistical tools)
    • Translate
    • Interpret
    • Read
    • Compute

Statistical Reasoning

  • Understanding
    • Explain why
    • Explain how

Statistical thinking

  • Apply
    • What methods to use in a specific situation
  • Critique
    • Comment and reflect on work of others
  • Evaluate
    • Assigning value to work
  • Generalize
    • What does variation mean in the large scheme of life

Empirical Cycle

By Adriaan de Groot

The components

  • Observation
    • Idea for hypothesis
  • Induction
    • General rule
    • Hypothesis
  • Deduction
    • Expectation / Prediction
    • Operationalization
  • Testing
    • Test hypothesis
    • Compare data to prediction
  • Evaluation
    • Interpret results in terms of hypothesis

Explained by Annemarie Zandscholten

Experiment

Video registration

Heads

bit.ly/2j54A2U

Emperical Cycle

  • Observation Patiënt is showing post traumatic symptoms
  • Induction Can we diagnose PTSD
  • Deduction H0: P: fair coin → C: patiënt is balanced
  • Deduction HA: P: Unfair coin → C: patiënt is unbalanced
  • Deduction HA: P: data ≠ EV → C: is unbalanced
  • Testing Choose α and Power
  • Evaluation Make a decision

Null distribution

Let’s analyse the null distribution of the results.

Google sheet

Distributions

What is the difference between

  • Population distribution
  • Sample distribution
  • Sampling distribution

Binomial distribution

(nk)pk(1−p)n−k

(nk)=n!k!(n−k)!

With values:

n = 10   # Sample size
k = 0:10 # Discrete probability space
p = .5   # Probability of head

Probabilities

Testing

I landed 2 times head. Can we conclude PTSD?

  • As you can see from the distribution of healthy coins, we cannot conclude that by definition.
  • What we can do is indicate how rare 2 is in a healthy population.

Testing

  • Based on the null distribution we can see that the expected value (EV is 5.)
  • We can now define the H0 hypothesis: H0=5
  • What is the alternative hypothesis?
  • The alternative hypothesis describes a situation where PTSD is pressent.
  • We could say that the alternative hypothesis is not 5.
    • HA≠5
  • We could also formulate our H0 and HA more abstract:
    • H0: the patient is balenced
    • HA: the patient is unbalenced
  • What criterium should we use to conclude that one would be unbalenced?
  • In the social sciences this α criteria is often 5%.
  • I tossed 2 times head. That is more frequent than 5%.
  • Therefore, we conclude that our patient is probably healthy but we can never be sure.
  • My coin could still be part of the unbalenced population.

Null distribution

End

Contact

  • @shklinkenberg
  • Klinkenberg
  • ln.AvU@grebneknilK.S
  • ShKlinkenberg

CC BY-NC-SA 4.0

SMCR / SMCO

1 / 27
Statistical Reasoning Null Hypothesis Testisn Klinkenberg University of Amsterdam 2024-02-05

  1. Slides

  2. Tools

  3. Close
  • Statistical Reasoning
  • Course structure
  • Course parts
  • in class assessment
  • Grading
  • Credits
  • Learning
  • Reasoning in statistics
  • Statistical Literacy
  • Statistical Reasoning
  • Statistical thinking
  • Empirical Cycle
  • By Adriaan de Groot
  • The components
  • Explained by Annemarie Zandscholten
  • Experiment
  • Heads
  • Emperical Cycle
  • Null distribution
  • Distributions
  • Binomial distribution
  • Probabilities
  • Testing
  • Testing
  • Null distribution
  • End
  • Contact
  • f Fullscreen
  • s Speaker View
  • o Slide Overview
  • e PDF Export Mode
  • ? Keyboard Help