Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Finding (and Listing) The Unlisted: A Strategy For Achieving A Listed Sample’s Cost Savings Without Sacrificing Coverage

Anthony Salvanto and Kathleen Frankovic, CBS News

Frankovic will be the presenter.

Contacts: salvantoa@cbsnews.com; kaf@cbsnews.com

Address for both: CBS News, 524 W. 57th St., New York, NY 10019

In pre-election polling, the lure of using registered voter lists is obvious: a shorter field time, more productive interviewing and lower costs. However, using lists results in non-coverage error that can be substantial and is often unknowable. CBS News developed a strategy for avoiding non-coverage while retaining many of the operational advantages of calling from registered voter lists. The paper will discuss the biases found in listed samples, including which types of voters and households are most likely to be omitted and how those biases impact survey design and results. The paper will report on several tests of the CBS News coverage strategy, using pre-election polls in South Dakota and South Carolina to examine costs, coverage and accuracy of pre-election estimates and will assess the possibility of utilizing this method in other circumstances.

Publication Type: Conference Paper/Unpublished Manuscript

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This paper reports the results of a pro bono study sponsored by Bernett Research, which provided 400 telephone interviews; Survey Sampling Inc., which provided a Random A random digit dial (RDD) sample; and QSA Research & Strategy, which provided design and analytic services. Survey results provide evidence that the do-not-call (DNC) list is working: nearly seven in ten respondents who had signed up for the registry said they are no longer receiving any telemarketing calls and more than nine in ten said they are receiving no calls or fewer calls. However, there is no evidence that DNC respondents receiving fewer telemarketing calls are any more likely to participate in surveys than they were before DNC went into effect. On average, respondents rated their likelihood of participation in future surveys at a “5” on a 10-point scale. This low rating seems to be due partly to lack of knowledge about the ethical standards of professional survey researchers and the ways that participation in surveys can benefit respondents as both consumers and citizens; but, it is also partly due to bad experiences with improperly-conducted surveys. Majorities of those who had been interviewed during the past 12 months reported that they had experienced an interview that was too long or was longer than they were told it would be. Majorities also said that they were asked questions that were poorly written, repetitive, or didn’t make sense to them. Although skip patterns are one of the most basic survey tools, a slight majority said they were asked detailed questions about topics they had already told the interviewer they didn’t know about. These findings suggest that the image of survey research suffers not only a failure to communicate the benefits of survey participation to the public but also from some survey researchers’ the lack of quality standards and consideration for respondents.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This paper will present the results of a statewide survey conducted in New Jersey in January of 2004. Two samples of 400 respondents each were conducted at the same time. One sample employed a common RDD sample; the other was drawn from a list of registered voters. This serves as the independent variable. Substantive dependent variables include field costs and interviewing statistics, political views, policy positions and demographics

Publication Type: Abstract
Abstract: Objective: To develop a comprehensive list of journals that are appropriate publication venues for SAdS scholars. Methods: An initial list of 127 journals was assembled by using previous journal-rating studies in SAdS, vitas from faculty and students at The University of Mississippi, Department of Pharmacy Administration (UM), and annual reports and publication lists on department SAdS websites where available. Current and former administrative leadership of the UM pharmacy administration department reviewed the list for comprehensiveness. A Delphi panel consisting of 15 faculty members representative of various SAdS subdisciplines was conducted online. They refined the list by assessing the appropriateness of each journal as a publication venue for SAdS scholars. Panelist demographics were collected. Results: Fourteen responding panelists agreed on the appropriateness of 13 journals. Additionally, the panel, at the close of the first contact, provided an additional 23 journal titles, which were submitted to the panel and assessed by 11 responding panelists during the second phase of the Delphi technique. The 134 journal titles which members did not reach consensus were submitted for reconsideration of journal title inclusion in a forthcoming instrument that would assess the perceived quality of SAdS journals. Seven panelists responded. Journals whereby >50% (a priori criterion) of panelists agreed that the journal should be in the subsequent phase of the study was included in the final list for a total of 113 journal titles. Implications: This list is being used in an instrument designed to measure deans’, faculty’s, and graduate students’ perceived quality of SAdS journals.