[ Spring 2010 ]
Update: Address-Based Sampling Nets Success for KnowledgePanel® Recruitment and Sample Representation
By Charles DiSogra
At Knowledge Networks our commitment to the highest quality online sample in the industry is paramount. We are happy to provide the latest update on our addressed-based sampling (ABS) efforts. This update illustrates how ABS delivers the online panel representation that you can count on for your research needs.
When Knowledge Networks introduced national addressed-based sampling in 2009 to recruit KnowledgePanel members, we knew it was the right decision at the right time. ABS is now exceeding our recruitment expectations for solving the diminishing efficiency of the former random-digit dial (RDD) landline telephone recruiting and delivering panel membership of previously hard to recruit groups .
Industry wide, RDD is delivering fewer and fewer young adults, fewer minorities and fewer low-education and low-income households. Also, by definition, RDD landline samples do not reach the nearly one out of four U.S. households that gave up their landline telephones to now be classified as cell phone only. Address-based sampling is designed to resolve these shortcomings and has done so successfully for KnowledgePanel® recruitment.
Positive ABS response trumps RDD telephone
First of all, KN's ABS mailed materials have consistently proven to be both effective and efficient. With national random samples of residential addresses, we are now getting about 14 out of 100 eligible households to positively respond to our mailed invitation by indicating an interest in joining KnowledgePanel. Of those, three out of four follow through and become participating panel members. In contrast, with the RDD telephone samples that used an interviewer-recruiter, only a little more than half of those expressing interest in joining would end up participating.
Most encouraging is that the KnowledgePanel ABS samples are delivering a demographically more balanced recruitment than is obtainable with telephone samples. Table 1 demonstrates the improved unweighted proportions of ABS-recruited panel members compared to the unweighted proportions of earlier RDD-recruited panel members. The percents for young adult, racial minority, Hispanic, low-education and low-income categories are all higher. The ABS percents are also more closely aligned with the percent estimates for the U.S. adult population.
As expected, we are seeing a healthy proportion of cell phone only households in the mix. Among all the recruited households that have contributed one or more new panel members, 17% are confirmed as being cell phone only households. The most recent estimates based on in-person national health surveys are that about 23% of U.S. households are cell phone only.
KnowledgePanel address-based recruitment samples are successfully delivering a statistically robust and highly representative membership profile. We are pleased to have this high quality, probability-based panel available for your important research projects.
Dr. Charles DiSogra is Knowledge Networks' Chief Statistician and heads the statistics unit at the Menlo Park office. He brings over 20 years experience in survey research, sample design, data analysis, and administration. Dr. DiSogra has a Masters degree in public health and a doctorate in nutritional epidemiology with an emphasis in biostatistics and policy analysis from the University of California at Berkeley.