Ch. 6



Null Hypothesis





Alternative/Research Hypothesis



Confound Hypothesis





Evaluating the Null Hypothesis



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...









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