Rectangular rapid flashing beacons (RRFBs) have proven to be a useful tool for improving driver yielding and pedestrians’ safety at midblock crossings. This study analyzed driver yielding at 23 RRFB-enhanced midblock crossings on three-lane roadways with and without median refuge islands in Oregon. The locations were chosen to represent a range of posted speed limits and average daily traffic that aligns with existing guidance for median and beacon installations. Sites were classified either as (a) no median refuge, RRFBs placed outside the roadway, (b) median refuge, RRFBs placed outside the roadway, (c) median refuge, RRFBs placed on the island and outside the roadway. Yielding was determined following protocols established in prior research. Two hundred seventy-six hours of video footage were analyzed, resulting in 3,065 crossing events (1,338 staged; 1,727 naturalistic) undertaken by 3,683 pedestrians. High yielding rates were observed—the average near side yielding rate was 97%, with the lowest site having a rate of 89.9%. Yielding rates were generally higher on the far side. Owing to sample size and consistently high yielding rates, it was not possible to make conclusive observations about the relationship between driver yielding and the presence of median or additional beacons for the volume and speed combinations. The results generally indicated that yielding rates increased with the addition of median beacons. The findings also suggested median refuge islands with a beacon increased yielding. The increase in yielding was statistically significant at sites with average daily traffic of 12,000 to 15,000. Over the last decade, many jurisdictions in the United States have installed pedestrian crossing enhancements (PCEs) to improve pedestrian safety and experience. These enhancements include continental markings, median refuge islands, curb bulb-outs, pedestrian-activated flashing beacons, overhead signs, advanced stop bars, and most recently, rectangular rapid flashing beacons (RRFBs). FHWA first issued Interim Approval (IA-11) for the use of RRFBs in 2008 (1), then rescinded it in December 2017 because of patent issues. However, on March 20, 2018, FHWA issued Interim Approval for the optional use of pedestrian-actuated rectangular rapid-flashing beacons at uncontrolled marked crosswalks (IA-21) after the patent issues were resolved. In recent years, several studies have been carried out to evaluate the effectiveness, measured by driver yielding rates, of RRFBs at midblock crossings (2–12). These studies have used either staged pedestrian crossings, naturalistic observations, or both to evaluate driver yielding behavior. A wide range of driver yielding rates have been observed, ranging from 19% to nearly 98%, though there are differences in how yielding is defined. Other studies have also evaluated the safety effectiveness of RRFBs and found crash modification factors of 0.53 and 0.71, indicating a significant reduction in pedestrian–motor vehicle crashes postinstallation (13, 14). Other features of enhanced crossings such as raised medians have been found to significantly lower pedestrian crash rates along multilane roadways at both marked and unmarked crosswalk locations (15). Other research has also demonstrated safety benefits for pedestrians from raised medians and refuge islands as individual enhancements (16–18). However, no studies have been found to evaluate the combined effect of median refuge islands and RRFBs. Since RRFBs are primarily used at midblock crossings, the first decision an agency often faces is whether to mark the crossing or not. Subsequent decisions for the agencies include the type of crossing enhancement to install and whether RRFBs are suitable for that location. The guidance set forth by FHWA presents the range of conditions for which RRFBs should be considered a treatment option. The guidance has number of lanes (two, three, four, or more), posted speed limit (≤30 mph, 35 mph, ≥40 mph) and vehicle volume (15,000 ADT) (19). This research aimed to explore driver yielding on three-lane roadways, with and without median refuge islands. This research observed driver yielding behavior in Oregon at 23 midblock crossings on three-lane roadways equipped with RRFBs with and without median refuge islands and beacons. Twelve hours of video footage was collected and coded at each site. Both staged pedestrian crossings and observations of naturalistic crossings were analyzed for pedestrian performance measures and drivers’ yielding behavior. The remainder of the paper is organized in the following manner. A description of data collection and coding is next, followed by the results and analysis, and finally, our conclusions.