Pavement smoothness is a key factor in determining highway user satisfaction. Because of technical limitations of traditional equipment and measurement procedures, the adequacy of a straightedge in characterizing roughness and having it relate to user response has come to question. One barrier to more widespread implementation of a new smoothness standard is the lack of objective, verifiable correlation methods for use in establishing specification limits. The purpose of this study was to (1) develop statistical approaches to harmonize two different measurement devices, i.e., a 3-m straightedge and an inertial profiler; and (2) determine the specification for flexible pavements. The device harmonization requires the development of procedures for statistically testing through comparisons to reference measurements. Data collected from 85 pavement sections on newly constructed and resurfaced highways were examined to demonstrate the device's ability to measure specific smoothness properties. Smoothness data are shown to be normally distributed according to the chi-squared test. Test results obtained from straightedge are found to be significantly different from those from profiler. Smoothness specifications applied for straightedge cannot be directly used to those for profiler. A conversion equation is developed to help contractors convert smoothness measurements between these two devices based on the same acceptance level. The International Roughness Index (IRI) obtained from the profiler appears to provide better measures of smoothness and better correlation with user response. Four following ranges are proposed for the smoothness specification according to the acceptance smoothness level: bonus, full pay, penalty, and correction. As the acceptance probability sets at 20%, the bonus threshold value should be less than 1.20 mm and 2.55 m/km for straightedge and profiler, respectively. The acceptance level is found to be below 2.60 mm and 3.60 m/km for straightedge and profiler, respectively. The demarcation line between penalty and correction is drawn at 95% of the acceptance probability with corresponding values of 3.35 mm and 4.10 m/km for straightedge and profiler, respectively. This research effort has developed a series of relationships between IRI and straightedge that can assist highway agencies in transitioning to IRI smoothness specification for pavements.