The decline in support has been driven by Republicans, who moved from 47 percent support in March to 30 percent now. Support has slipped among independents, but by a smaller degree. Two-thirds of Democrats currently support the deal, as do half of independents.
Republican lawmakers and presidential candidates have been railing against the deal for months, but Senate Democrats have recently twice blocked efforts to kill the deal. Republicans are now trying a different tack by voting Thursday to stop Obama from lifting sanctions on Iran.
A variety of public polls have shown declining support for the deal, falling as low as 21 percent approval in a Pew Research Center poll earlier this month (with 49 percent disapproving). Much of the variation in support or opposition between different polls can be traced to how the questions have been asked. Questions that include more details about the deal have generally found more support, while questions with fewer details find less support.
The Post-ABC poll tested the deal with two questions asked of random half-samples. One question used similar language to that in three other Post-ABC surveys going back to November 2013, describing the lifting of economic sanctions against Iran in exchange for Iran agreeing not to produce nuclear weapons. In this poll and another in July, the question noted that sanctions could "snap back" if Iran broke the agreement. This description found 51 percent in support and 41 percent opposed.
The alternative question asked of the other half-sample excluded the description of inspections or the potential for sanctions snapping back sanctions and found a more narrow divide of 45 percent support to 44 percent opposition.
But the trend line is clear across all polls, and that is that the deal is becoming less popular.
The six-point gap in support between the two questions (51 percent for the full question, 45 percent for the shorter version) was not statistically significant given the sample sizes. Across demographic groups, the gap in support between question wordings is relatively consistent, for which support was 10 points lower in the question that did not mention the potential for sanctions to snap back in place.
The Post-ABC poll was conducted Sept. 7-10 among a random national sample of 1,003 adults, including landline and cellphone respondents. Results among the half samples of 483 and 515 respondents each have a margin of sampling error of plus or minus 5 percentage points.