Know for Quiz #4 on A&A chapters 3-6:


1. Be able to look at a scatterplot and

(a) distinguish between a strong and weak, positive and negative correlation.

(b) identify any outliers (outriders)

2. Know that Pearson's r measures linear association among quantitative variables (interval or ratio scales).

Factorial ANOVA:

1. Understand the differences between one-way and factorial ANOVA.

2. Be able to look at a graph and a table of means and tell

(a) if there is an interaction effect (are lines parallel?)

(b) if there are main effects (are marginal means different?)

(Of course, to see if any effects are significant, you need to run the ANOVA in SPSS.)


1. Understand the basic logic of the chi-square: You have categorical data-counts of people/events/objects that fall into different categories. You are comparing observed frequencies (how many people in each category for your sample) with the frequencies that would be expected under the null hypothesis.

2. Like the F ratio, the chi-square is always positive. Like the F distribution, the chi-square distribution is positively skewed (skewed to the right). Unlike the F distribution, the expected value if Ho is true is close to 0 (not close to 1)

3. Know that there are three types of chi-squares (two goodness of fit-"equal proportions" and "compare," plus chi-square for independence. The differences in the three are basically

(a) they address different questions

(b) you go through different steps to generate the expected frequencies

(c) there's a different formula for df for GF (C-1 for both kinds) and independence (C-1)(R-1)

4. Know the restrictions: Independent observations, expected frequencies of at least 5 for each cell.

NOTE: Both Factorial ANOVA and Chi-square involve tables with numbers in them. What's different is that for Factorial ANOVA, the numbers in the cells are MEANS of quantitative variables. For Chi-square, what's in the boxes is FREQUENCY COUNTS (either observed or expected).