The presidential race between Jimmy Carter and Ronald Reagan is much closer than recent, highly publicized public opinion polls have suggested, according to an analysis of national polls by The Washington Post.
As the Democratic delegates began arriving in New York for tomorrow's national convention, the backers of President Carter were apparently confident enough of victory to concede some rules and platform issues to the main challenger, Sen. Edward M. Kennedy (D-Mass.), in the interest of conciliation and party unity. [Details on Page A5.]
A new round of polls being released today and later this week will show Republican nominee Reagan holding a lead ranging from 14 points to 20 points over President Carter.
That is a sharp decrease from a 27-point lead reported by pollster Louis Harris two weeks ago. At this current rate, Reagan's substantial lead could easily disappear by November.
A new Gallup poll, for release today, shows Reagan leading in a three-way race with support from 45 percent of registered voters. Carter second at 31 percent and independent candidate John B. Anderson trailin at 14 percent. That poll was conducted Aug. 1 through 3, before Carter's press conference last Monday in which the president spoke to the nation about his brother Billy's relationship with the radical Arab nation of Libya.
Last night, CBS reported new poll findings, saying that Carter had made at least temporary gains after that press conference. The network's polling immediately before the press conference showed Carter 23 points behind Reagan. When polling after the press conference was included, the deficit had decreased to 20 points.
Harris also polled immediately before and after the president's press conference last Monday, and he said that his findings, which will be released later in the week, show Reagan with a 19-point lead over Carter -- 8 points closer than the period just after the Republican National Convention.
In addition, Harris includes more Republicans and fewer Democrats in his polling samples than do other major polling organizations, making his findings more favorable to the Republicans.
Virtually all of the major national polls showed Reagan gaining in popularity after his nomination in Detroit, but the outstanding political question is whether Reagan's lead over the president is so substantial that it cannot be overcome.
The Associated Press-NBC poll yesterday reported a margin of 25 percentage point, but noted that 51 percent of the people interviewed said they had not made up their minds.
The NBC-AP survey did find that Carter is the preferred candidate for the Democratic nomination among rank-and-file Democrats, leading Kennedy, 38 to 21 percent.
The new poll reports could have impact on the Democratic National Convention, starting tomorrow in New York, influencing the last ditch drive to deny the party's nomination to Carter. The anti-Carter movement has been fueled by the claim that Carter is a sure loser in November.
Using poll findings to buttress their case, many Democrats have urged Carter delegates to desert him, saying that he stands to take a great many House and Senate Democrats down with him.
The findings most often cited by the anti-Carter activists are those of Harris, and for good reason: Far more than any other pollster, Harris has for the past year or so been suggesting disaster for the Democrats if they nominate Carter.
Last week, a Harris poll report said that "there is evidence of a dramatic shift in the basic party identification of the voters," and maintained that the Republicans could well take control of Congress in the Nov. 4 election.
But Harris's findings, which have been coming out more frequently than others and drawing more attention, are by their very nature more likely to give higher marks to Reagan and the Republicans than are other polls.
The conclusions Harris draws came frome samples that have far more Republicans in them than samples drawn by CBS and the Gallup organization.
Harris's sample of 1,195 people last week looked like this: 39 percent were Democrats, 32 percent were Republicans and 26 percent were independents. Harris called these people "likely voters."
He started out with far more people than 1,195, and with a higher proportion of Democrats and a lower proportion of Republicans. But after weeding out and disregarding the views of those he said he thought were not likely to vote, he ended up with a sample that had in it only 7 percentage points more Democrats than Republicans.
CBS also aimed for a "likely voter" sample in its polling, using a complicated formula to produce what Cathy Frankovic of the network yesterday referred to as a "final probable electorate." That sample was 43 percent Democratic, 26 percent Republican and 31 percent independent.
In the CBS sample, then, Democrats outranked Republicans by 17 percentage points, compared to 7 percent in the Harris poll. As might be expected, the CBS poll showed Carter trailing Reagan by a smaller margin than did the Harris poll.
The Gallup organization's poll made no attempt to sort out likely voters, instead reporting results based on interviews with a cross-section of 938 people who said they were registered to vote. Its sample: 47 percent Democratic, 27 percent Republican and 24 percent independent -- or 20 percentage points more for the Democrats than the Republicans.
Just how sharply the composition of those polled affects results of a poll may be seen by applying the Harris technique to someone else's poll findings.
For example, during July, at the time of the Republican convention, the Roper organization, using its own "certain voter" filter, found Reagan ahead of Carter by 34 to 31 percent in a three-way-race, with Anderson drawing 18 percent, and 17 percent saying they were uncertain as to how they would vote.
The makeup of Roper's sample was 52 percent Democratic, 27 percent Republican and 20 percent independent.
Burns Roper, chairman of the Roper organization, said that if he used the same sampling technique as Harris, Reagan would have been shown holding an 11-point lead over Carter instead of a 3-point lead.
Roper said his own sample may have slightly over-represented Democrats. Whether or not it did, polling by Roper in a three-month period has shown none of the shift toward Republican affiliation and away from the Democrats that Harris cited. The same is true of the Gallup findings over the last few months.
Roper, in fact, expressed astonishment at Harris' reported trend. Putting his own findings aside, Roper said that Gallup's registered voter figures are "much closer to what a voter group would be" than are Harris'.