This project will be the FINAL PHASE of a genre of NITC-sponsored research that includes previous NITC contracts 547, 650, 763. and 1103. As background, contracts 547 and 650 were used to build station area databases for 12 light rail transit (LRT) systems, nine bus rapid transit (BRT) systems, four streetcar transit (SCT), and five commuter rail transit (CRT) systems. For contract 763, we evaluated 13 BRT systems. For all three studies, we included employment data extending annually from 2002 through 2011, and census data for 2000 and 2010. Our analysis compared development, demographic and housing outcomes associated with these transit systems during the period before the Great Recession (2000 through 2007) and during recession into recovery (2008 through 2011). We also conducted limited hedonic regression analyses using very large distance bands from stations or corridors based on CoStar commercial rent data from the early 2010s. Though we found important differences in development outcomes between the study periods of 2000-2007 and 2008-2011, our research could not measure or compare outcomes during the period of economic stability that commenced since the practical end of the Great Recession, which was 2012. We thus updated and analyzed those databases to 2015 with contract 1103. This allowed us to compare outcomes associated with these transit systems before (2000-2007), during (2008-2011) and after (2012-2015) the Great Recession. Moreover, since many systems did not start until during or after the Great Recession, the updated database allowed for more comprehensive assessment of outcomes between the Great Recession and early recovery into the period of relative economic stability. Through technology transfer, contract 1103 allowed us to create a database for researchers, students and policy analysts. The FINAL PHASE of this genre of research will disaggregate data used to assess systemwide outcomes to outcomes based on types or clusters of stations by mode. Our proposed research will be guided by two overarching research questions. The FIRST QUESTION is: How do Transit Development Outcomes Vary by Mode and Type of Transit Station? This analysis will include each transit system for each metropolitan area studied during appropriate time periods for that system, as well as systems combined across metros. Trends to be tested include: (1) Changes in the number and share of jobs by sector with respect to type of system and distance from stations, by type of station based on factor/cluster analysis; (2) Changes in the number and share of jobs by wage category with respect to type of system and distance from stations by type of station based on factor/cluster analysis; and (3) Changes in number and share of population, households, householders by age, and housing by tenure with respect to type of system and distance from stations by type of station based on factor/cluster analysis. The SECOND QUESTION is: How does the real estate market for office, retail and apartment properties respond to proximity to transit stations by mode and type of station? Our prior work pioneered the use of CoStar commercial rental data for very broad assessments of real estate market responsiveness to transit by type but not really by location except for corridor distance bands. The proposed research will conduct more refined relationships in those metropolitan areas based on mode and type of transit station where CoStar data are sufficient for analysis. In addition, we will update our complete database with a codebook for anyone to access through NITC. We intend this to be a resource that advances research into and supports policy decisions related to LRT/BRT/SCT/CRT systems planning across the nation. In all, our FINAL PHASE will consider 18 LRT systems, 16 BRT systems,12 SCT systems and seven CRT systems serving 36 metropolitan areas. More importantly, the FINAL PHASE will allow us to use factor/cluster analysis to create typologies of station areas to assess the extent to which types of stations (as opposed to transit systems as a whole) make a difference in economic development (based on LEHD data), and people (census data) during the periods before, during and after the Great Recession as appropriate for each system and mode. It will also allow us to refine hedonic regression analysis. The FINAL PHASE will add new material to the database that will also be shared through technology transfer. Our METHODOLOGY AND ANTICIPATED RESULTS is based on two approaches. In the FIRST APPROACH, we will use ECONOMIC BASE METHODS to assess change in concentration in: (1) Jobs by sector; (2) Jobs by wage category; (3) Population and households by age and other demographic features; and (4) Residential units and tenure relative to central counties with respect to transit mode by time period (before, during and after the Great Recession). Using FACTOR/CLUSTER analysis, we will refine analysis to consider outcomes with respect to types of transit stations by mode. Factor/cluster analysis dimensions will include the following among others as appropriate: (1) Land use mix (an entropy measure); (2) Jobs-population balance (a measure of jobs versus population concentration); (3) Distance to downtown and other major activity centers (a centrality measure); (4) Employment sector composition (a measure of economic concentration); and (5) Socioeconomic composition (a measure of demographic concentration). Without predicting those types, we anticipate they include combinations of spatial (downtown, suburban center, isolated), economic (high-middle-low wage corresponding to relative education and skill levels), and social dimensions (age-income-race/ethnicity-tenure). We anticipate that there may be surprises compared to our prior work, such as gains in concentration at types of transit stations that otherwise saw losses of concentration for systems as a whole, and vice versa. The SECOND APPROACH will be based on HEDONIC REGRESSION ANALYSIS using CoStar data to assess the association between transit station distance and rent with respect to different system modes, station types, and metropolitan areas. We will further advance prior work in another way: Assessing the interactive effect of transit station type and distance on commercial properties within discrete distance bands of stations. In our current work (not yet published), we have discovered that commercial markets express segmentation effects based on distance to transit stations, by mode. For instance, commercial property rents are much more responsive to BRT station proximity than to SCT proximity for reasons that are different between these systems. In these ways, the proposal will allow us to explore these refinements to market responses to transit systems by mode, type (based on factor/cluster analysis) and distance. Disaggregating our earlier and ongoing research to assess outcomes based on types/clusters of stations is a logical and often requested extension of our NITC-sponsored research. The FINAL PHASE allows us to do so.