THE RACE BETWEEN HILLARY CLINTON AND DONALD TRUMP IS WITHIN THE MARGIN OF ERROR IN WISCONSIN. CHARLES FRANKLIN IS THE DIRECTOR OF THE MARQUETTE LAW SCHOOL POLL AND JOINS US NOW. THANKS FOR BEING HERE.

CHARLES FRANKLIN:

THANK YOU.

ZAC SCHULTZ:

WELL, AFTER A BIG POST-CONVENTION BUMP FOR HILLARY CLINTON, THE LAST POLL SHOWED THINGS SETTLING AND THIS SHOWS THE SAME.

CHARLES FRANKLIN:

IT DOES. IN FACT IT SHOWS VERY LITTLE MOVEMENT. IT’S EITHER A TWO-POINT OR THREE-POINT CLINTON LEAD WITH LIKELY VOTERS, DEPENDING ON WHETHER YOU INCLUDE LIBERTARIAN OR GREEN CANDIDATES. AND IT WAS A THREE-POINT RACE THREE WEEKS AGO. SO REALLY WE’VE SEEN VIRTUALLY NO CHANGE IN THE PRESIDENTIAL RACE. IN CONTRAST TO THE POST-CONVENTION BOUNCE BACK IN AUGUST THAT CLINTON REALLY BENEFITED FROM, WHICH IS NOW PRETTY MUCH DISSIPATED.

ZAC SCHULTZ:

AND WHEN YOU DO ADD THOSE THIRD-PARTY CANDIDATES, YOU SEE THE SUPPORT, BUT IT DOESN’T CHANGE THE OVERALL LEAD.

CHARLES FRANKLIN:

ONE OF THE INTERESTING THINGS IS THIRD-PARTY CANDIDATES IN THE PRESIDENTIAL RACE AND ACTUALLY ALSO IN THE SENATE RACE DON’T SHIFT THE NUMBERS VERY MUCH. REALLY IT’S A POINT EITHER WAY. IT’S ROUNDING ERROR, IN EFFECT, AT THIS POINT. EVEN THOUGH THE THIRD-PARTY CANDIDATES ARE GETTING A BIGGER SHARE OF THE VOTE THAN THEY NORMALLY DO.

ZAC SCHULTZ:

LET’S TALK ABOUT THAT BECAUSE TYPICALLY A THIRD-PARTY CANDIDATE, THAT WILL EVAPORATE BY ELECTION DAY AS PEOPLE DECIDE TO MAKE A CHOICE AT THE TOP OF THE BALLOT. WHEN WILL WE SEE THAT TAKE PLACE IN THE POLLS, OR WILL WE?

CHARLES FRANKLIN:

WHEN WE LOOK AT HISTORICAL POLLING THAT HAS THIRD PARTY CANDIDATES IN MOST OF THE TIME, THEY COME DOWN, BUT THEY’RE STILL A BIT OVERESTIMATED IN THE LAST POLLS BEFORE ELECTION DAY. LIBERTARIANS HERE FOR EXAMPLE AVERAGE ABOUT 1% OR LESS IN ACTUAL ELECTION RESULTS IN PRESIDENTIAL RACES. GARY JOHNSON IS POLLING A LOT HIGHER THAN THAT. BUT THE QUESTION IS IS HE GOING TO BE LIKE ROSS PEROT IN ’92, WHO HAS REAL STAYING POWER, MAYBE BECAUSE OF DISCONTENT WITH THE PRESIDENTIAL NOMINEES OR DO WE SEE HIM FOLLOW MORE THE HISTORICAL PATH AND TREND DOWN? WE DON’T KNOW AT THIS POINT WHICH IT’LL BE.

ZAC SCHULTZ:

TALK ABOUT DISSATISFACTION WITH THE CANDIDATES. BOTH TRUMP AND CLINTON HAVE A LOT OF PEOPLE THAT SAY THEY’RE NOT THE PREFERRED CHOICE FOR THEIR NOMINEE. 68% OF REPUBLICANS. 48% OF DEMOCRATS. AND THEN WHEN YOU ASK WHO THOSE PEOPLE WHO ARE DISSATISFIED, WHERE THEY’RE VOTING, THAT’S WHERE YOU SEE A LOT OF THIRD-PARTY SUPPORT.

CHARLES FRANKLIN:

ABSOLUTELY. IT’S WORTH NOTING THAT BOTH OF THESE NOMINEES LOST THEIR RESPECTIVE PRIMARIES HERE, SO MAYBE IT’S NOT SURPRISING THAT A LOT OF PEOPLE SAY I WISH I HAD SOMEBODY ELSE. BUT PARTY LOYALTY AMONG THAT GROUP OF REPUBLICANS OR DEMOCRATS IS DOWN AT 60 OR 55. IT’S NOT NEARLY THE 85 TO 95 THAT WE WOULD NORMALLY EXPECT. AND GARY JOHNSON DOES QUITE WELL WITH BOTH GROUPS. JILL STEIN, THE GREEN CANDIDATE, DOES BETTER WITH DEMOCRATS AND NOT SO WELL WITH REPUBLICANS. BUT THEN THERE’S ALSO STILL A FAIR NUMBER OF UNDECIDED. WHAT THIS MEANS IS THERE’S SOMETHING CLOSE TO 20% OF THE VOTE THAT’S NOT GOING TO THE TOP TWO NOMINEES THAT MAYBE IS UP FOR GRABS, POSSIBLY THE PRESIDENTIAL DEBATE WILL HELP FOCUS AND MAY OR MAY NOT SHIFT THAT.

ZAC SCHULTZ:

MOVING ON TO THE U.S. SENATE RACE, RUSS FEINGOLD STILL HAS A LEAD OVER REPUBLICAN RON JOHNSON. THAT HASN’T CHANGED MUCH.

CHARLES FRANKLIN:

NOT MUCH. IT WAS THREE POINTS LAST TIME. IT’S FIVE POINTS NOW. IN A THREE CANDIDATE RACE, IT’S A SIX-POINT LEAD IN THE TWO CANDIDATE HEAD-TO-HEAD. MAYBE A LITTLE BIT UP. BUT THE WAY TO CHARACTERIZE IT I WOULD SAY IS FROM LOW SINGLE DIGITS TO MID SINGLE DIGITS.

ZAC SCHULTZ:

FEINGOLD LEADS IN SOME OTHER CATEGORIES THAT ARE INTERESTING. NOT ONLY DOES HE HAVE A BETTER FAVORABILITY RATINGS, BUT MORE VOTERS CALL HIM SOMEONE WHO CARES ABOUT ME AND HE’S REMEMBERED AS A MORE EFFECTIVE SENATOR. DID WE JUST EXPLAIN WHY HE’S IN THE LEAD?

CHARLES FRANKLIN:

I THINK WE LARGELY DID. SENATOR JOHNSON HAS BOUNCED AROUND A LITTLE MORE FAVORABLE THAN UNFAVORABLE IN THIS POINT. TWO POINTS MORE UNFAVORABLE. BUT HE’S BEEN PRETTY STEADY THERE. FEINGOLD HAS HAD A PRETTY SUBSTANTIAL MORE FAVORABLE THAN NOT. AND CARES ABOUT YOU, WHICH WE’RE MEASURING NOW, HE HAS AN ADVANTAGE. AND A SMALLER ADVANTAGE ON THE EFFECTIVENESS. YOU CAN SEE IN SOME OF JOHNSON’S RECENT ADS THAT ARE AIMED AT SHOWING CARING AND CONCERN FOR CONSTITUENTS. FEINGOLD HAS HAD THOSE ADS FOR A WHILE.

ZAC SCHULTZ:

YOU HAD A VERY INTERESTING QUESTION ABOUT THE PERCEPTION OF VOTER FRAUD OUT THERE OR THE AMOUNT OF PEOPLE THAT MAY BE INTIMIDATED AWAY FROM VOTING. WHAT’S THE OVERALL TAKE-AWAY?

CHARLES FRANKLIN:

I THINK THAT WE’VE SEEN IN PAST POLLING THAT THERE’S A LOT OF SUPPORT FOR PHOTO I.D. BILLS IN THE STATE. THAT’S BEEN CONSISTENT. BUT THIS TIME WE ASKED PEOPLE HOW MANY PEOPLE WILL VOTE ILLEGALLY? AND HOW MANY PEOPLE WHO ARE LEGAL CITIZENS WITH A RIGHT TO VOTE WILL BE PREVENTED FROM VOTING? AND DEPENDING ON HOW YOU MEASURE IT, THERE’S A PLURALITY THAT THINK IT WILL BE LESS THAN 100 PEOPLE VOTING ILLEGALLY AND IT’S 57% THAT THINK IT WILL BE MORE THAN 100 PEOPLE WHO WILL BE PREVENTED FROM VOTING. AND WHEN WE LOOK AT THE TWO DIRECTLY, WE FIND THAT IT’S TWO TO ONE. 18% THINK THERE WILL BE MORE ILLEGAL BALLOTS CAST. 37% THINK MORE PEOPLE WILL BE DISENFRANCHISED. THIS IS OF COURSE A POLICY DEBATE IN THE LEGISLATURE AND IT’S ALSO A SUBJECT OF COURT ACTION. BUT PUBLIC OPINION HERE IS SEEING MORE PEOPLE BEING DISENFRANCHISED THAN ILLEGAL BALLOTS CAST.

ZAC SCHULTZ:

IN THE LAST POLL THE MAKEUP OF THE PARTICIPANTS WAS A LITTLE MORE REPUBLICAN THAN USUAL. I SAW EVEN SOME CONSERVATIVE COMMENTATORS SAYING THAT MAY EXPLAIN WHY THAT HAPPENED IN THE LAST POLL. SAME THING THIS TIME?

CHARLES FRANKLIN:

SAME THING THIS TIME. REALLY SINCE THE CONVENTIONS WE’VE SEEN DEMOCRATS MAINTAIN THEIR CONSISTENT HISTORIC LEVEL OF SUPPORT. BUT A LITTLE BLIP UP AMONG REPUBLICANS BY TWO OR THREE POINTS IS NOT HUGE, BUT IT’S ENOUGH TO CLOSE MUCH OF THE GAP BETWEEN DEMS AND REPS. AND IT’S ALMOST ALL COMING FROM INDEPENDENTS. IT LOOKS LIKE MAYBE THERE’S SOME INDEPENDENTS THAT REALLY DIDN’T THINK OF THEMSELVES AS REPUBLICANS BEFORE, BUT SUPPORTING DONALD TRUMP MAY BE CALLING THEMSELVES REPUBLICANS NOW AND THAT’S BUMPING UP THAT NUMBER A LITTLE BIT. BUT NOTICE THAT THE VOTES MARGINS DIDN’T MOVE IN THE DIRECTION OF THAT SHIFT, SO IT DOESN’T SEEM THAT IT’S REALLY AFFECTING THE VOTE SO MUCH AS MAYBE WHAT PEOPLE ARE CALLING THEMSELVES.

ZAC SCHULTZ:

I WANT TO GET TO AN EXPERIMENT YOU PARTICIPATED IN RECENTLY. THE NEW YORK TIMES AND SIENNA UNIVERSITY CONDUCTED A POLL IN FLORIDA AND GAVE THE SAME RAW DATA TO FOUR OTHER POLLSTERS. YOU WERE ONE OF THEM AND THEY ALL CAME UP WITH DIFFERENT RESULTS. HOW DID THAT HAPPEN?

CHARLES FRANKLIN:

FIRST OF ALL, CREDIT TO THE NEW YORK TIMES FOR DOING THIS EXPERIMENT AND TO SIENNA FOR BEING WILLING TO LET THEIR DATA GO OUT TO US. POLLSTERS ALWAYS MAKE DECISIONS ABOUT WHAT IS THE UNIVERSE OF VOTERS THAT YOU’RE GOING TO CARE ABOUT, WHAT YOU’RE GOING TO PREDICT TO, AND WHAT’S THE DEMOGRAPHIC MAKEUP OF THAT ELECTORATE. AND THAT’S A MATTER OF PROFESSIONAL JUDGMENT. THERE’S NOT ONE ABSOLUTELY CLEARLY RIGHT ANSWER FOR THOSE THINGS. SO THEY GAVE THE FOUR OF US THESE THINGS AND SAID GO TO IT. NOW, THEY EMPHASIZED THAT THE RESULTS VARIED QUITE A BIT. OF THE FOUR OF US THAT DID CONVENTIONAL POLLING, WE GOT TWO WITH A ONE-POINT CLINTON LEAD, MINE WAS A THREE-POINT AND THERE WAS ONE WAS A FOUR-POINT LEAD. I LOOK AT THAT AND SAY THERE’S NOT A WHOLE LOT OF RANGE THERE. BUT THERE WAS A NEW FANCY STATISTICAL MODEL, WHICH IS VERY DIFFERENT FROM WHAT THE REST OF US DID, THAT SHOWED A ONE-POINT TRUMP LEAD. SO THAT’S A RANGE. FIVE POINT RANGE IN THIS SAME DATA.

ZAC SCHULTZ:

A LOT OF THAT DEPENDS ON HOW YOU DECIDE WHO’S A LIKELY VOTER, MEANING WHO WILL ACTUALLY SHOW UP ON ELECTION DAY.

CHARLES FRANKLIN:

AND HERE THERE ARE TWO BIG DIFFERENCES USED IN THIS SURVEY. I AND ONE OF THE OTHERS USED PEOPLE’S SELF-DESCRIPTION OF WHETHER THEY’RE GOING TO VOTE. YOU KNOW BETTER THAN ANYBODY WHETHER YOU’RE GOING TO VOTE THIS TIME. YOU TELL ME. OTHERS AND A LOT OF PARTISAN POLLSTERS DO THIS, LOOK AT THE VOTE RECORDS FROM THE ACTUAL VOTER REGISTRATION LIST AND SEE HOW MANY TIMES IN THE LAST THREE OR FOUR ELECTIONS DID YOU VOTE AND USE THAT TO PREDICT RATHER THAN YOUR TELLING ME WHAT YOU’RE GOING TO DO. THERE ARE ADVANTAGES AND DISADVANTAGES TO BOTH OF THESE METHODS. WE, THE MARQUETTE POLL, DON’T USE VOTER LISTS TO BEGIN WITH, SO WE DON’T EVEN HAVE THAT OPTION. BUT MANY POLLSTERS AND A LOT OF PARTY POLLSTERS DO. I THINK THE RANGE THAT YOU SAW WAS PARTLY RELATED TO HOW DID YOU DO IT, USING THE SELF-REPORT WAS THREE POINTS FOR MINE AND FOUR POINTS FOR ONE OF THE OTHERS THAT USED SELF-REPORT. THOSE THAT USED THE VOTER LIST SHOWED A LITTLE TIGHTER RACE, A ONE-POINT CLINTON LEAD.

ZAC SCHULTZ:

THAT SHOULD LET PEOPLE KNOW WHEN THEY’RE LOOKING AT THESE WIDE DISCREPANCIES OF POLLS FOR DIFFERENT STATES AND THE NATIONAL POLLING, THAT CAN EXPLAIN WHY THEY COULD BE SO DIFFERENT.

CHARLES FRANKLIN:

EXACTLY. AND IT ALSO REALLY WARNS YOU ABOUT WHY NOT TO MAKE APPLES TO ORANGES COMPARISONS. WHEN YOU’RE COMPARING ONE POLL BY ONE POLLSTER WITH ANOTHER POLL BY ANOTHER, WE MAY HAVE MADE DIFFERENT JUDGMENTS ABOUT WHAT THE SHAPE OF THE ELECTORATE, MAKE UP OF THE ELECTORATE WOULD BE AND WHO’S A LIKELY VOTER. BUT WHEN YOU COMPARE THE SAME POLLSTER OVER TIME, YOU LOOK AT THE MARQUETTE POLL FOR EXAMPLE, WE’RE DOING EVERYTHING EXACTLY THE SAME AND WE DECIDED HOW TO DO THAT BEFORE JANUARY. AND SO NOTHING CHANGES IN OUR METHODOLOGY. SO YOU’RE MAKING AN APPLES TO APPLES COMPARISON THAT WAY.

ZAC SCHULTZ:

THAT’S WHY WE CALL YOU THE GOLD STANDARD. THANKS FOR YOUR TIME TODAY.

CHARLES FRANKLIN:

THANK YOU.