2007 AAPOR Presentations

Knowledge Networks adheres to the AAPOR Code of Professional Ethics and Practices.

RESULTS OF A WITHIN-PANEL SURVEY EXPERIMENT OF DATA COLLECTION MODE EFFECTS USING THE GENERAL SOCIAL SURVEY'S NATIONAL PRIORITY BATTERY
J. Michael Dennis, Knowledge Networks; Email
Rick Li, Knowledge Networks; Email
Joe Hadfield, Knowledge Networks

This study is an attempt to contribute to previous research on the subject of data collection mode effects comparing specifically the Internet mode of data collection to telephone-based and in-person data collection. In this study, we controlled for sample source by having all interviews conducted with pre-recruited panelists from KnowledgePanel®. The administered survey questions are the "national priority" items from the General Social Survey (GSS). The actual GSS was in the field for the in-person main study during the fielding of our experiment. In the analysis, we compared the results from the three modes of data collection – web, telephone, and in-person – to identify categories of questions where the mode of data collection is related to a directional difference in the survey findings.

More Info - PowerPoint | Full Paper

AN INVESTIGATION OF PANEL CONDITIONING WITH ATTITUDES TOWARD U.S. FOREIGN POLICY
Poom Nukulkij, Knowledge Networks; Email
Joe Hadfield, Knowledge Networks
Stefan Subias, Knowledge Networks; Email
Evan Lewis, Program on International Policy Attitudes (PIPA); Email

While the advantages of survey panels are well-known, one potential challenge in managing a survey panel is the possibility of panel conditioning. The concern is that the way participants respond to survey questions is influenced by prior survey taking.

We investigate the matter using KnowledgePanel®, the RDD-recruited Internet panel managed by Knowledge Networks. Members of KnowledgePanel® complete approximately one survey per week and continue participating as long as they choose. Our first question is whether panelist tenure is predictive of attitudes toward U.S. foreign policy. The survey data come from a September 2006 survey conducted for the Program on International Policy Attitudes at the University of Maryland.

Our second question is whether prior participation on a survey topic relates to attitudes on the subject. The diverse mix of survey topics administered to members of KnowledgePanel® presents one major difference between longitudinal studies that are centered on a specific subject. In those circumstances, participants know what kinds of questions to expect or what kinds of behaviors they will be asked about. With KnowledgePanel®, foreign policy is a regular survey topic presented to panelists. In the past four years, Knowledge Networks conducted 50 separate surveys about foreign policy. This volume allows an examination of whether prior participation on the same subject relates to attitude measures.

Our initial analyses do not find evidence of panel conditioning. In our regression models, neither panel tenure nor participation in prior foreign policy surveys significantly predicted Presidential job approval ratings, isolationism versus global view, satisfaction with position of the United States in the world, diplomacy versus military use, and preferences for amount of defense spending.

More Info - PowerPoint | Full Paper

A COMPARISON OF RESULTS FROM AN ALCOHOL SURVEY OF A PRE-RECRUITED INTERNET PANEL AND THE NATIONAL EPIDEMIOLOGICAL SURVEY ON ALCOHOL AND RELATED CONDITIONS
Timothy Heeren, Boston University
Erika M. Edwards, Boston University
J. Michael Dennis, Knowledge Networks; Email
Sergei Rodkin, Knowledge Networks; Email

Youth Alcohol Prevention Center, Boston University School of Public Health and Knowledge Networks (KN) conducted a national survey of 3,000 respondents between the ages of 18 and 39 who had consumed alcohol at some point during their lives. The study was funded by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). The results from this KN survey were compared to the current "gold standard" face-to-face survey of alcohol-related behaviors and problems, the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Estimates of drinking characteristics based on KnowledgePanel® reasonably reflect those of the national population of drinkers, with KnowledgePanel® representing a somewhat higher level of drinking and risky drinking. Compared to NESARC, KnowledgePanel® reported somewhat higher levels of drinking behavior. Multiple logistic regression analyses were performed using KnowledgePanel® and NESARC samples, modeling the odds of a lifetime diagnosis of alcohol dependence on age, gender, race/ethnicity, education, family history of alcohol problems, age of drinking onset, and risky drinking during the period of heaviest drinking. Results from both samples show no significant association between race/ethnicity or education and dependence; both samples show similar significant increases in the odds of dependence for males, those with a family history of alcohol problems, those with earlier age of drinking onset, and those with moderate or risky drinking patterns.

More Info - PowerPoint