Our first hypothesis was that the proportions of classes with different levels of homework are equal. We concluded that this hypothesis was consistent with the data, 2 (2, N = 101) = 0.55, p = .7579. Our second hypothesis was that the level of homework varies across grade categories. This hypothesis was also consistent with the data, 2 (4, N = 101) = 51.75, p = < .001, = .72, Cramer's V = .51, = .79. To explore the association further, we looked at the ability to predict Grade from Homework ( = .30) and the ability to predict Homework from Grade ( = .42). Overall, we concluded that less homework is typically assigned in lower grades.

	How many classes were included in the study?
	What is the largest marginal frequency for a grade level category?
	What is the level of measurement of the two variables (N = nominal, O = ordinal, I = interval, R = ratio)
	If 101 classes are divided evenly across homework levels, what would be the expected frequency within each homework level?
	What is the value of k for the one-variable chi-square test?
	What are the degrees of freedom for the two-variable chi-square test?
	If the null hypothesis of the two-variable test were true, what would be the value of phi?
	Is the association positive (P), negative (N), or near zero (Z)?
	If the researchers had accidentally arranged "Heavy" between "Light" and "Moderate", which of the measures of association would you expect to be most affected? (P = phi, V = Cramer's V, L = lambda, G = gamma)
	In the case of the one-variable test, would you reject (R) or fail to reject (F) the null hypothesis?
	What proportion of 10 to 12 grade classes have Heavy homework?
	What proportion of 1 to 6 grade classes have Heavy homework?