Instructions_for_Ç2_TEST - 2 TEST To do 2 you...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
χ 2 TEST To do χ 2 you MUST have: 1. DISCRETE , NOMINAL ( CATEGORICAL ) DATA 2. IF ONLY 2 CATEGORIES , NO EXPECTED VALUE < 5 IF MORE , NO MORE THAN 20% OF EXPECTED VALUES < 5, AND NONE <1 χ 2 = Σ (Observed Value - Expected Value) 2 / Expected Value χ 2 Goodness of fit test Most of you learned about this test in biology classes when you tested whether genotype frequencies were in Hardy-Weinberg equilibrium. This was is sometimes called a chi- square test of goodness of fit, where you are asking, do the numbers of observations occur as often as expected given (or “do they fit”) some pre-existing hypothesis of expected relative frequencies. For example, your hypothesis might be that people like dogs, cats, mice and fish equally well as pets. You’d survey 100 pet-owners and tally the numbers of times each category of pets was represented. If your categories are as below, then expected value = (A+B+C+D)/4 = sum of all cells/ #
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 08/28/2011 for the course ESE 585 taught by Professor Richardhoward during the Fall '11 term at Purdue University-West Lafayette.

Ask a homework question - tutors are online