RLI 288  10 RESEARCH LIBRARY ISSUES: A REPORT FROM ARL, CNI, AND SPARC 2016 In another common networked-service technique that will increase the validity of the sample, the technology group at the university campus can push out the appropriate library access links (open URL link resolver) within Google Scholar to browsers on campus, thus increasing traffic to the link resolver and the proxy server if one is used, tightening up the web of possible survey points. The most comprehensive interception point is at the Internet service provider router for the university, and as radical as it may seem, this router-based solution has been implemented twice by one university and has been contemplated by several others. The router-based solution has the fewest limitations, but the other intercept points also work well, and collect reliable samples, as long as the limitations are understood. University of Massachusetts Amherst Environment and Implementation The University of Massachusetts (UMass) Amherst is a public research university offering undergraduate, graduate, and professional degrees. There are 28,518 undergraduate and graduate students, and 1,170 full- time instructional faculty. The UMass Amherst Libraries is the largest state-supported academic library in New England with more than 8 million items, spending $6 million on continuing e-resources. The UMass Amherst Libraries implemented MINES twice, once using the two-hour random moment sample design in 2008–2009, and then the every Nth sample design in 2013–2014. The first survey ran for 12 months, from September 2008 to August 2009 with two two-hour survey periods per month. Figure 1 is a screen shot of the survey. FIGURE 1. SCREEN SHOT OF MINES FOR LIBRARIES SURVEY AT UMASS AMHERST