I am working with a client on LEED EBOM 2009 project located in an urban area. The building has over 2,051 FTEs and multiple tenants. We don't have access to email addresses for each of the occupants, but do have tenant contacts who can forward information to the tenants. Given the location, we believe more than 75% of the occupants commute via some form other than a high-emitting single-occupancy vehicle. Our challenge is reaching the 80% response rate in order to extrapolate the data for the whole building.

We are looking at conducting the survey using the random sampling approach per the reference guide. However, since we don't have access to tenant's individual names and email addresses, we can't randomly select and contact tenants. We may be able to ask our tenant contacts to distribute the survey to a random group of their employees, however we then have less control of the distribution and return. Alternatively, we could position a team at the elevators and sample every fourth person who enters - although it would be random, we're not sure it would be a truly representative sample of the building population.

Any thoughts on these two methodologies, or others that would be acceptable to the reviewer that we should consider, to achieve the highest return using the random sample approach?