This project requires the use of the HSB N=300 dataset.

1. Plot the frequency distributions or stem-and-leaf displays for the achievement variables READING, MATH, and SCIENCE, and for the social/personality variables LOCUS, MOTIVAITON, and SELF-CONCEPT. What would you say about the shapes of the distributions for the achievement variables and the social/personality variables. Which variable seems to have the greatest skew?

2. An official of the agency that funded data collection for High School and Beyond observed: "Boys performed significantly better on standardized tests than girls. This has serious policy implications in how the needs of boys and girls are addressed in American schools." Compare the performance of boys and girls on reading, writing, math, and science measures. Is this a defensible statement?

3. The of the three social and personality variables, LOCUS, MOTIVATION, and SELF-CONCEPT, which is the best predictor of reading achievement? Run a regression predicting reading achievement from that variable. What is the regression coefficient, b? Please state in words what b tells you about reading scores.

4. Using Analysis of Variance, compare the group means on READING for different SES levels. Are the three SES groups significantly different from each other, using the protected t-test? Calculate e (see p. 239 of Welkowitz) for SES. What do these analyses allow you to conclude about the role of SES on reading achievement?

Due date: December 7, 5:00 PM. Please put your projects in Kenji Hakuta’s mailbox in CERAS; you may also submit your project in Powerpoint, MS Word, or a webpage format.