2009 JSM PRESENTATIONS
COMPUTING RESPONSE RATES FOR PROBABILITY-BASED WEB PANELS
Mario Callegaro, Knowledge Networks; Email
Charles A. DiSogra, Knowledge Networks; Email
This paper builds and advances a previous publication on how to compute response metrics for online panels (Callegaro & DiSogra, 2008). It starts by describing how multi-stage response rates are computed for probability-based web panels (recruitment, profile, and completion rate) and their use in a cumulative response rate calculation. We then focus on standing panels that rely on continuous recruitment from multiple independent recruitment samples and the concept of recruitment cohort. We will also present a special case for studies that use a screener and refinements that take panel attrition rates into account. Lastly it is discussed how to combine mixed mode recruitment methods in the computation of the recruitment rate.
More Info – PowerPoint | Full Paper
RECRUITING PROBABILITY-BASED WEB PANEL MEMBERS USING AN ADDRESS-BASED SAMPLE FRAME: RESULTS FROM A PILOT STUDY CONDUCTED BY KNOWLEDGE NETWORKS
Charles A. DiSogra, Knowledge Networks; Email
Mario Callegaro, Knowledge Networks; Email
Erlina Hendarwan, Knowledge Networks, Email
Address-based sample (ABS) frames have emerged as a solution to shrinking RDD frame coverage due to cell-phone-only households. Knowledge Networks' (KN) probability-based on-line KnowledgePanel® has used RDD frames for telephone recruiting since 1999. In 2008, KN conducted a pilot study (n= 9,994) to recruit panelists via mail using an ABS frame. KN developed and tested recruitment materials (advance postcard, initial mailer, a mailer to non-responders), and a mailing and telephone protocol. Half were mailed the advance postcard, half were not. Randomized portions of each half received a $5 cash incentive, $1, or $0. Non-responders with a landline match were called; the rest mailed a final letter. Response was allowed via mail, online, or toll-free number. Results are presented showing demographic response, mode response, nested incentive conditions and advance postcard effects.
More Info – PowerPoint | Full Paper
