Page 250 - Solid Waste Analysis and Minimization a Systems Approach

228 DATA COLLECTION

99 SIC code groups. Surveys that were missing information, such as company name or
number of employees were cross-referenced with a number designated to each survey.
Once the company was determined from this survey number, the number of employees
was collected from the Harris Registry and corporate information from various Internet
sources (www.cnn.money.com). Surveys that were incomplete were discarded
(6.8 percent or 30 surveys were discarded). The following is a summary of the figures:

■ A total of 4135 businesses were contacted and requested to participate.
■ Only 1500 companies (36.3 percent of those contacted) verbally agreed to partici-
pate in the survey.
■ A total of 468 companies (31.2 percent of those who agreed to participate) returned
the survey.
■ Only 438 (29.2 percent of those who agreed to participate) companies correctly
returned the survey with useable data.

Of the 4135 companies initially contacted, 438 correctly returned the survey, result-
ing in an overall response rate of 10.6 percent. This is a very common response rate
for a “cold mailing,” in which random surveys are mailed without initially contacting
potential respondents. The use of phone calls, e-mails, and faxes before mailing the
survey aided to reduce the percentage of nonresponders. Prescreening potential
respondents with phone calls, e-mails, or faxes increased the response rate of the
actual mailed survey by nearly 20 percent (from 10.6 to 29.2 percent). The pre-screening
process eliminated companies that would definitely not respond to the survey and
identified a pool of potential companies that would be more likely to complete the sur-
vey, hence increasing the response rate. The number of surveys mailed and success-
fully returned from each SIC code group can be seen in Fig. 14.2.
Waste material generation given in terms cubic yards was converted into annual ton-
nage using the average conversion factors developed by the EPA. An example of a
conversion is discussed in Sec. 8.6. This method was applied to convert all generation
rates given in volume to weight.
Potential sources of error in the data collected are

■ Inaccurate data generated from the surveys (including the solid waste management
district surveys and the national survey).
■ Data collected may not accurately describe the population. Companies that are more
conscientious about environmental initiatives and protection may be more likely to
submit results.
■ Some respondents had difficulty separating facility financial data from corporate
data, but made attempts to do so. Surveys with incomplete financial data were dis-
carded or the company was contacted and asked for an estimate.
■ Some respondents were unwilling to release financial data (including annual waste
costs) on the survey. Some of the existing data did not include these figures either.
Landfill disposal charges in the area were used as a substitute. This information was
easily attainable from the companies and various Internet sources, including the
Solid Waste Association of North America (www.swana.org).