1: How Does Binary Forced Choice vs Offering a "No Opinion" Choice Influence Survey Results?
2: How Have Opinions Changed as a Result of Events in the News?

859 adults nationwide were interviewed by SurveyUSA Monday 01/07/13 through Wednesday 01/10/13. Research, conceptualized and commissioned by iMediaEthics.org, was conducted 100% by telephone, as follows: Adults reachable on a home telephone (67% of respondents) were interviewed on their home telephone in the recorded voice of a professional announcer. The youngest-male method of respondent selection was used on these landline calls. Respondents unreachable on a home telephone (aka: "cell phone respondents")(33% of respondents), were contacted on their cell phone by live operators, who hand-dialed the telephone, remained on the line throughout the duration of the call, read the questions to respondents, and recorded the answers. Blended sample for both landlines and cellphones provided by SSI of Fairfield CT. Responses were minimally weighted to U.S. Census targets for gender, age, race and region. Each respondent was randomly assigned at the time of the interview to one of two different paths through the survey; depending on the path, respondents heard slightly different question wording. Where further necessary, questions and answer choices were rotated to prevent order bias, recency and latency effects. SurveyUSA assigns to each question within the instrument a theoretical margin of sampling error, but such error is useful only in theory, and though commonly cited in the presentation of research results, sampling error is only one of many types of error that may influence the outcome of an opinion research study. More practical concerns include the inability to contact some, the refusal of others to be interviewed and the inability of still others to speak the language. It is difficult to quantify non-sampling errors.