Methodology Statement and Final Topline Results
Methodology Statement by KnowledgePanel (KP) OmniWeb
FULL REPORT HERE
OMNIWEB using the KnowledgePanel ™ is a national online omnibus service of GfK Custom Research North America. The KnowledgePanel™ is the only commercially available online probability panel in the marketplace; making the sample truly projectable to the US population, which sets it apart from traditional “opt-in” or “convenience” panels.
- To recruit panel members, we use ABS (address-based sampling) as the primary methodology — a probability-based approach that includes cell phone-only households
- For non-Internet households, we provide Internet access and a laptop for the completion of online surveys: Therefore, the sample is not limited to only those who already have Internet access
- The representativeness of its sample — including hard-to-reach groups such as young adults, males and minorities — has been documented in numerous academic papers
The results contained in this report are based on interviews of U.S. adults, 18 years and older, conducted in two separate surveys: November 22-24, 2013, and separately December 5-8, 2013. All completed interviews are weighted to ensure accurate and reliable representation of the total population, 18 years and older.
The raw data are weighted by a custom designed computer program, which automatically develops a weighting factor for each respondent. This procedure employs several variables, including: age, sex, education, race, HH income, met/non-met status, internet status and geographic region. Each interview is assigned a single weight derived from the relationship between the actual proportion of the population with its specific combination of age, sex, education, race and geographic characteristics and the proportion in our sample that week. Tabular results show both weighted and unweighted bases for these demographic variables.
The November survey includes 1,026 interviews, while the December survey includes 1,031 interviews. For each survey, the margin of error on weighted data is + 3 percentage points for the full sample.
Final Topline Results (Compiled by iMediaEthics)
Interviewing dates: November 22-24, 2013 and December 6-8, 2013
Note: Due to rounding, percentages may not add to 100. An asterisk (*) indicates values less than 0.5%.
FORM A QUESTIONS ASKED OF HALF THE SAMPLE
A1. Here is a policy that has been considered by some cities and states around the country. Please tell me if you would favor or oppose it: Prohibiting restaurants from using trans fats in foods
Dec Nov
6-8 22-24
65% 64% Favor
33 33 Oppose
2 3 No opinion
A1a. How upset would you be if restaurants were NOT prohibited from using trans fats in foods? (Asked of respondents who favor policy.)
Dec Nov
6-8 22-24
9% 5% Very upset
31 23 Somewhat upset
46 51 Not too upset
14 21 Not upset at all
0 * No opinion
A1b. How upset would you be if restaurants WERE prohibited from using trans fats in foods? (Asked of respondents who oppose poliicy.)
Dec Nov
6-8 22-24
13% 15% Very upset
28 21 Somewhat upset
40 44 Not too upset
20 19 Not upset at all
* 1 No opinion
COMBINED RESULTS FROM A1, A1A AND A1B:
Here is a policy that has been considered by some cities and states around the country. Please tell me if you would favor or oppose it:
Prohibiting restaurants from using trans fats in foods
Dec Nov
6-8 22-24
26 18% Favor – upset if not prohibited
39 46 Favor – not upset if not prohibited
2 3 No Opinion
20 21 Oppose – not upset if prohibited
13 12 Oppose – upset if prohibited
FULL REPORT: GfK Web-Based Polls: Why Do AP, Pew, CBS, and Other Media Trust Them? A Recent GfK Poll
GfK responds to iMediaEthics’ results, iMediaEthics responds to its claims