This thesis investigates the relationship between place attitudes and measurable place attributes in Santa Barbara, CA. People's relationships to places form a key component of travel behavior and decision making, but they have been the subject of limited empirical study. In addition to improving our understanding of place attitudes, I investigate appropriate ways to model a cross-classified dataset with spatially autocorrelated ordinal responses. Data sources include a spatially constrained place attitudes survey, parcel-based land use, business establishment records, and a collection of local geotagged Tweets, from which a spatial measure of happiness was extracted. The relationships among these variables are investigated using a cross-classified multilevel ordinal regression model, which finds several significant relationships between place attitudes and available opportunities, land use, and natural amenities. I also find that while Twitter data may provide another way to measure place attitudes, more work is needed to develop useful variables from it.