Surveys based on samples of customers are a common element in many evaluations and research projects. Surveys can be administered by mail, phone or e-mail. Sometimes they include on-site visits to gather data or install loggers. The more extensive the data collection effort is, the more likely it can benefit from a real-time survey management system that is Web-enabled.

Here is a typical example of how a Web-enabled Survey Management System (SMS) can be used to improve the survey process. The example is a case where loggers have to be installed in businesses for six weeks, and then picked up later. Customers must be contacted by phone first and give their agreement to have the loggers installed in their business.

First the SMS will read in all of the population data, condition it as needed, and select the sample frame given the parameters of the sample design. Often the sample design will include proportional stratification over several variables to help ensure an unbiased sample. For example, there may be quotas by geography, size of customer (based on energy use) and business type (based on NAICS code).

As the survey process starts, the SMS will put up randomized list of customers and phone numbers from the sample frame. Callers will start making phone calls down the list of names. Because the SMS is real-time Web-enabled, multiple callers can work at the same time on the same set of data with no overlap in calls. Each call attempt is tracked in the system. When a customer is reached and has responded 'yes' or 'no' to participating in the logger study, their name will be removed from the remaining list of names to call.

This is where the Strat-o-matic comes in. As customers respond 'yes' to the invitation to participate, they are counted in real-time against the required quotas for the stratification variables. If a particular quota cell is full, say suburban medium-sized grocery stores, then all remaining customers in that cell in the sample frame are removed from the remaining call list. As time goes on, calls will be concentrated on the particular cells that still need to be filled. The Strat-o-matic ensures that no cells will be over-filled.

It is most efficient to arrange appointments for home visits immediately after a customer has agreed to participate, while they are still on the phone. The SMS is linked to a scheduling system and a mapping system. The mapping system allows the caller to see the location of the customer they are talking to and the location of other participants that will be receiving visits and the time of those visits. With this information in hand, the caller can make efficient appointment schedules for the installer.

Use of the SMS then shifts from the callers to the installers. Every day, installers can print out a map and a list of directions to get to all of their appointments for the day. After they have visited a home, they can record logger numbers and any information they collected in the home into the SMS via any Web connection. Later, the SMS can be used to plan pick-up routes to collect the loggers.

Not all of these components are in every survey, but the SMS can be quickly customized to any survey effort. The most time-intensive part of any new SMS development is usually the initial effort to obtain and condition the population data. It is also possible to let an established survey house use an SMS in tandem with their own CATI system. The SMS can be used for generating call lists and creating schedules and maps while the CATI system is used to gather data from a set of survey questions. Projects that used Web-enabled Survey Management Systems built by Daniel are listed on this page.

Pepco Holdings Direct Load Control Study (2010 to 2011)

Commonwealth Edison Residential CFL Study of Illinois and Kansas, including automatic creation of individual cover sheets to go out in the field (2010)

Puget Sound Energy Residential HVAC Duct System Evaluation (2010)

Tucson Electric Power Residential and Commercial DSM Baseline and Potential Study (2010)

PECO Energy DSM Baseline Study (2010)

Arizona Public Service Commercial & Industrial Lighting Study (2010)

Arizona Public Service Residential HVAC Quality Inspection Study (2010)

AEP-Ohio DSM Baseline and Potential Study (2010)

CPUC Low Income Solar Inspection Project (2010)

Progress Energy-Carolinas Energy Efficiency Benchmarking (2010)

Commonwealth Edison Residential HVAC Evaluation (2009-2010)

Natural Resources Canada Net to Gross Study, including analysis of results using logit and probit models. (2009-2010)

Ottertail Power DSM Baseline and Potential Study (2009)

Palm Desert HVAC Study, including Commercial Refrigerant Charge and Airflow (RCA) sites. (2009)

California Public Utility Commission High Impact Measure C&I Lighting Study, including creating of a centralized inventory system for loggers. (2009)

Nova Scotia Potential Study (2009)

Arizona Public Service Residential Lighting Study, including code to access the raw logger hexidecimal data and extract internal logger settings (2009)

California Public Utilities Commission Residential Gas Study (2008-2009)

Minnesota Office of Energy Security DSM Potential Study (2008-2009)

Union/Enbridge Commercial Free-rider and Spillover Survey (2008)

Northwestern Energy Home Energy Audit Evaluation (2008)