Ch. 6

Null Hypothesis

Alternative/Research Hypothesis

Confound Hypothesis

Evaluating the Null Hypothesis

• Differences in Variability

• Types of variation



• Systematic variation





• Chance variation







• Between groups variation











• Within groups variation

So, if the null hypothesis is the best explanation, then...

If the alternative is the best explanation, then...

Statistically Evaluating the Null Hypothesis

Step 1:

Compute the probability that the treatment effect is due to chance alone

Step 2: If the probability that the treatment effect is due to chance is low enough...

Step 3: If the probability that the treatment effect is too high...

• how low is low enough?

Errors in Science

True state of affairs (reality)

Decision

Correct decision	Type II Power
Type I -limited to alpha	correct decision

Threats to internal validity

History

Maturation

Testing

Instrumentation

Statistical regression

Selection

Mortality

Selection-Maturation Interaction

Diffusion/Imitation of treatments