This dissertation compares and analyzes the relationship of design-build (DB) and design-bid-build (DBB) project delivery methods with performance metrics of large highway projects. Project performance is measured in terms of cost, schedule, safety, change orders, and quality on these two types of highway projects. The performance benchmarking methodology used here is derived from work done on a Texas Department of Transportation (TX DOT) study of the State Highway (SH) 130 Project. Because SH 130 is the first DB highway project in Texas and is being built under a new contractual concept called the Comprehensive Development Agreement (CDA), this dissertation establishes a framework for evaluating the performance of large DB highway projects. The CDA approach is an innovative form of the DB project delivery method that allows the contractor to simultaneously undertake right-of-way acquisition, utility adjustment, design, and construction activities. Because this approach is being used for the first time on a state highway project in Texas, it is beneficial to track highway project performance in order to assess whether this project delivery method is a better alternative for building high priority highways. The main objective of this dissertation is thus to compare the performance of large recent DB highway projects (in the context of SH 130) with similar on-going instate DBB highway projects. The research hypothesis is to determine whether there is a statistical difference in mean performance between DB and DBB highway projects. For large, recently built DB highway projects (Federal Highway Administration, Special Experimental Project Number 14 & Cost > US $ 100 million) and four of the largest, most recently built in-state DBB highway projects are identified for comparison purposes. This dissertation provides a detailed methodology to collect data and gives the results of performance benchmarking of these large DB and DBB highway projects. It also investigates associations or relationships between project characteristics (input variables) and project performance (output variables) of large highway projects. While previous analyses of DB and DBB methods have included a wide range of construction projects as varied as buildings and industrial facilities, this dissertation isolates the analysis of these two delivery approaches for large highway projects. It also helps to develop a method to collect data for benchmarking of large highway projects. This research should help TX DOT choose the appropriate delivery method for large future highway projects. / text