Threat of bias has kept many from using data gathered in less than optimal conditions. We maintain that when convenience sampling represents race and gender at nearly correct proportions and can be beneficial, as these two variables are quite often used as stratification variables. We compared a convenience sample with a proven sample. Race and Sex were nearly proportional as was found in the proven sample. We conclude that the convenience sample can be used as though it is simple random.
Thatcher, W. Gregory and Drane, J. Wanzer
"On Treating A Survey Of Convenience Sample As A Simple Random Sample,"
Journal of Modern Applied Statistical Methods:
2, Article 21.
Available at: http://digitalcommons.wayne.edu/jmasm/vol2/iss2/21