Online polls with self-selected respondents can be ridiculously non-reflective of the population as a whole. The results often depend on how effective respondents are at getting their friends on social media to respond to some news outlet's Big Question of the Day.
This NBC News/SurveyMonkey poll, however, looks somewhat legit.
The sample size for this week's poll is more than 10,000, which is pretty solid. Unfortunately, the sample population skews heavily white and relatively affluent. See the Methodology explanation and demographic breakdowns on the last page of the embedded PDF.
You'll have to scroll to page 5 of the document to get to the one question in which Gary Johnson and Jill Stein are mentioned. The poll, which has been collecting results since January, added this four-way-race question only last week.
This NBC News/SurveyMonkey poll, however, looks somewhat legit.
The sample size for this week's poll is more than 10,000, which is pretty solid. Unfortunately, the sample population skews heavily white and relatively affluent. See the Methodology explanation and demographic breakdowns on the last page of the embedded PDF.
You'll have to scroll to page 5 of the document to get to the one question in which Gary Johnson and Jill Stein are mentioned. The poll, which has been collecting results since January, added this four-way-race question only last week.
Johnson's 9% showing is fairly consistent with the Quinnipiac and Monmouth polls recently released.
Beyond the inclusion of Johnson and Stein, in itself worth celebrating, we see yet again:
Beyond the inclusion of Johnson and Stein, in itself worth celebrating, we see yet again:
- the numbers showing that Bernie Sanders does better than Hillary Clinton in a two-race with Donald Trump (page 4), and
- the almost consistently larger percentages in the Independent and Neither Party columns than found in the Republican and Democrat columns (pages 2 and 3).
RSS Feed
