The Washington Post - USA (2020-11-22)

(Antfer) #1

SUNDAY,NOVEMBER 22 , 2020 .THEWASHINGTONPOST EZ RE A25


SUNDAY Opinion


T


he firsttime Imet Barack
Obama,hestruckmeasdifferent
from anyother politicianIhad
ever met. He was smart, well-
read,affableandenergetic,butthatisn’t
whatmadehimstandout.Itwastheway
he asked questions. Most politicians ask
aquestiontoansweritthemselves.After
giving you abrief opportunitytore-
spond,theyjumpin,“Well,here’s whatI
think ...”and proceedto deliversome
packaged piece of wisdomtheyhave
doubtlessreciteddozens of times. But
Obamawouldaskaquestiontowhichhe
actually wantedananswer.Hewould
listen and ask anotherquestion. He
genuinelywantedtounderstand how
someoneelsemightviewanissue.
That unusual politiciancomes
throughclearly in his newbook, “A
PromisedLand.”Itiswell written, cer-
tainly the best-written presidential
memoirIhaveread. Obamahas an easy
and stylish waywithwords.Describing
walkingthroughthe West Colonnade of
the White House, he says, “it was where
each morningIfelt the firstslap of
winter wind or pulse of summerheat.”
Describingahelicopterride,hewrites,“I
gazed out at the rolling Maryland land-
scapeandthetidyneighborhoodsbelow,
and then the Potomac,glistening be-
neaththefadingsun.”
Themostnotable feature of the book,
however,isObama’sabilitytosee not
justbothsidesofeveryissuebutevento
empathize with the side in vigorous
oppositionto his own. He writes thathe
could understand HillaryClinton’s frus-
tration,afteralongclimbtopower,tobe
confrontinganupstartchallengerforthe
Democratic nomination. He under-
stands the motivations of Republican
leaderssuchasJohnBoehnerandMitch
McConnelland providesashorthistory
lesson:“Americanvoters rarely reward
the oppositionfor cooperating with the
governingparty.” He even has a“grudg-
ingrespect” forthe waythe tea party
gained passionate supportand wide-
spreadnewscoverage.
This qualityoffairmindednessis ad-
mirable in anyone, especiallyone who
has risen to the top of acutthroat
profession likepolitics. And it gave
Obamaconsiderable advantages in both
domestic and foreignpolicy. He could
see the world with different people’s
eyes, which broadenedhis horizonsand
made him abetternegotiator.But his
memoir does have one gap, alacuna in
his vision, both as president and as a
writer.Hedevoteslittletimeinthebook
to the central politicaldynamicin his
years in office —the rise of an enraged,
utterly obstructionist, Manichean oppo-
sition to his presidency, and himself
personally,thateventuallyculminatedin
theelectionofDonaldTrump.
Areminder: Barack Obamawas a
moderateDemocrat —“conservative in
temperament,”heacknowledges —and
governedas one. Forhis keyeconomic
advisers, he chose the mostcentrist,
market-friendly experts of the party,
Lawrence Summersand Timothy
Geithner.Hekept on GeorgeW.Bush’s
defense secretaryand offered another
keyCabinetjob, commercesecretary, to
RepublicanSen. Judd Gregg. He sent in
thousandsmore troops to Afghanistan
and expandeddrone warfare. And his
health-careplan was modeledon the
conservative HeritageFoundation’s old
proposal,one that also served as the
basis for Mitt Romney’sprogram when
hewasgovernorofMassachusetts.
Thisreignofmoderationandcompro-
mise, however,elicited areaction from
the RepublicanParty thatwas furious
and vengeful. Gregg, who initially ac-
cepted the job as commercesecretary,
had to back down in the face of activist
outragethathewas serving the enemy.
Obama recounts the case of Charlie
Crist, who as governor of Floridasup-
ported Obama’sstimulus,whichthe
statedesperately needed becauseits
economywasinfreefall.Histwo-second
handshakeand hug with Obamamade
himsotoxicwithintheRepublicanParty
thatby2010hehad to become an
independentandlateraDemocrat.
Despitemanycompromises, Obama
gotnot oneRepublican vote for his
stimulus or health-care bills in the
House of Representatives. And opposi-
tion to his policieswas often couchedin
blatantly racistways, such as posters
denouncingObamacarewithcaricatures
ofhimasanAfricanwitchdoctorwitha
bone stuck throughhis nose. Theman
who succeededhim in office, Trump,
rose to politicalprominencebycasting
doubt on whether Obamawas born in
theUnitedStates.
Obamatalks about these hysterical
reactionstohimintelligentlybutbriefly,
never offering deep analysisor passion-
ateanger.Headmits he wasn’t focused
ontheominousundercurrentsthatwere
growingin strength. “Myteam and I
were too busy,” he writes.But it might
also be thatitwould take him into deep
and dark waters thatare so different
from the hopeful, optimistic countryhe
so plainlywants to believe in. America
remainsforhimapromisedland.
[email protected]

FAREEDZAKARIA

What


Obama’s


memoir


leaves out


BYMADALINSAMMONS

P


residentTrump has lostreelec-
tion, and now Democratshavea
shot at taking back the Senate
withtworunoffelectionsinGeor-
gia. Asaborn and bred Southernerwith
more than afew Democratic campaigns
under my belt, Ihavesome advice for
out-of-staters eager to help the Demo-
craticcandidatesinGeorgia:Stayhome.
If there is anything the party should
have learnedin the pastfour years, it’s
thatwhenDemocratstrytonationalizea
competitive race, the effortoften back-
fires. It happenedearlierthismonth:
That’s whySusanCollinsisstillaRepub-
licansenatorfromMaineandwhyDem-
ocratsAmy McGrathinKentuckyand
Jaime Harrison in South Carolinawon’t
begoingtotheSenate.
In the 2018 midtermelections, when
recapturingthe House was Democrats’
top priority, Isaw firsthand how the
nationalize-the-campaign strategycan
fail. Iwas workingone of those races
deemed“red to blue”bythe Democratic
CongressionalCampaignCommittee.
ThenationalsupportforRichardOjeda’s
run was overwhelming.People from

California,Massachusetts,NewYorkand
morefilledoursmallRepublicandistrict
in West Virginia. Yetoncetheywent to
work, manyvoters regardedthem as
invadingstrangerswhohadnoideahow
totalktosomeonefromAppalachia.
TheDemocraticbluewaveofthe2018
midtermswasn’t as big as it could have
been, when several of the House races
thatpeoplewereexcitedabout,including
oursinWestVirginia,cameupshort.
NowI’m afraid Democratswillmake
the same mistakeinthe Georgia runoffs
on Jan. 5between RepublicanSen. Kelly
Loeffler and Democrat Raphael War-
nock,andRepublicanSen.DavidPerdue
and Democrat JonOssoff. Floodingthe
statewithwhatfeelslikeenthusiasmand
supportcould be aroadblockfor the
Democraticorganizersandfieldworkers
whohavededicatedthemselvesforyears
to this effort. Former Georgia congress-
woman and gubernatorial candidate
StaceyAbrams in particular has moved
mountainsto give these Democratsa
chanceofwinning.
Before an armyofnational celebrities
and Twitter warriorsswarm Georgia,
theyshouldkeepinmindthattheydon’t
knowthestate,thepeopleorthepolitical

dynamics.Everyone justwants to help,
but imagine a19-year-old Oregonian
spendingherChristmasbreakfromYale
Universityknockingondoorsandtrying
to tell someonein Atlantahow to vote.
Imagine having your phone ring every
night with people apparently calling
from Boston or Brooklyn explaining
what’sbestforGeorgia.
It’s counterproductive. And it could
spellthedifferencebetweenmoreWash-
ington gridlockand aCongressthatthe
Biden administration can work with to
putthenationonabetterpath.
Out-of-state Democratswho want to
help would do better to supportGeorgia
fromhome.DonatedirectlytotheOssoff
and Warnock campaignsto ensure they
have the resourcesneededto compete
withtheinevitabledelugeofRepublican
TVads.DonatetoFairFighttomakesure
organizersare supported. Donate to the
Democratic PartyofGeorgia as it gears
up for whatpromisesto be twoofthe
mostclosely watched Senate races in
recent memory. But awordofcaution:
Evenanavalancheofcampaignspending
canhavedrawbacksifalltheTVadvertis-
ing and mailedfliers and phone calls
starttoseem likeawretchedlyexcessive

outsideattempttobuyanelection.
Democratswere understandablyec-
static to see ablue Georgia on the
electoralmap,buttheywouldbewiseto
rememberthatalthoughmanyRepubli-
cans in Georgia casttheir presidential
vote for Democrat JoeBiden, theyalso
voted Republicandown-ballot—includ-
ingfor the Senate. Ablue Georgia was a
rebukeofTrump, not an embrace of
Democratic control.Democratsacross
the nation will want to push a“flip the
Senate”message, but theyshould letthe
Ossoffand Warnock campaignsrun on
the messagethatgot them this far—a
messageabout affordablehealth care, a
living wage, and protecting Social Secu-
rityandMedicare.
Don’t risktwo potentialSenate seats
forthesakeofagoodhashtag.Anddon’t
go to the Peach State unless organizers
ask for help. Youcan getaselfie with
StaceyAbramssomeothertime.

Thewriter, seniorcommunicationsadviser
at United Public Affairs, is the former
communicationsdirectorfor Andrew Yang’s
2020 presidentialcampaignand Richard
Ojeda’s2018WestVirginia congressional
run.

Democrats, please think twice before invading Georgia


weeksleadinguptothepresidentialelec-
tion. To complete 1,510interviews over
severalweeks,wehadtocall136,688vot-
ers.Inhard-to-interviewFlorida,only1in
90-oddvoterswouldspeakwithourinter-
viewers. Most calls to voters went unan-
sweredor rolled overtoansweringma-
chines or voicemail,never to be inter-
vieweddespitemultipleattempts.
Thefinal wave of polling,conducted
Oct. 25-27tocomplete500 interviews,
wasthe worstfor cooperation. We could
finishinterviewswithonlyfour-tenthsof
one percentfrom our pool of potential
respondents. As aresult, this supposed
“randomsamplesurvey”seeminglyyield-
ed, as did mostall Floridapolls, lower
supportfor PresidentTrump than he
earnedonElectionDay.
After the election, Inoted wide varia-
tionsincompletionratesacrossdifferent
categories of voters,but nearly all were
stilltoolowforanyactualrandomnessto
beassumedorimplied.
Manyvoterswhofitthe“LikelyTrump
Supporter”profilewerenotwillingtodo
an interview. It was especially hard to
interviewolder men. Similarly,wewere
less likely to complete interviews with
TrumphouseholdsinMiami’smediamar-
ket.Whateverthemotivation,thisbehav-
ior almostcertainly introduced bias into
pollresults,dampeningapparentsupport
forTrump.
Pollsters and poll readers can antici-
patelowandvariablecooperationratesto
persist,underminingrandomness.Inan-
ticipation of this,cooperation rates need
to be publishedwith all polls, to add a
dashofreal-worldsobrietytoourweigh-
ing of poll results.Presently,thisisvery
rarelydoneforpublicorprivatepolitical
polling.Ifyoudon’tbelieveme,askyour
pollster for his “disposition of sample”
reportand getreadyfor some cagey
equivocation.
Somesayonlinepollingwillhelp,andit
may. But mostonlinepolling uses non-
random samples from pre-recruited
“panels”of voters who have signed up to
be interviewed,typically for some incen-
tive.Andonlinesurveyshaveseriousdata
qualityorintegrityissues. Most voters
rush through them too rapidly for real
thought.Andwecannotverifythatonline
votersare indeedregistered to vote or
have the requisitevote historytheymay
claim.
One promising approachto making
online samplesmore verifiable and ran-
domisbytextinginterviewrequeststoa
genuinelyrandomsampleofthoseonthe
voter rolls. But standingpatonthe old
ways,ordenyingthenon-randomnessof
today’spolls, won’t makepolls great
again.

ThewriterisdirectorofHillResearch
Consultants and a2020fellow at the
University of SouthernCalifornia’sDornsife
Centerfor the PoliticalFuture.

BYDAVIDHILL

T


here’s adirty little secretthatwe
pollsters need to own up to: Peo-
pledon’ttalktousanymore,and
it’smakingpollinglessreliable.
WhenIfirstundertooktelephonepoll-
ingintheearly1980s,Icouldstartwitha
cluster of five demographicallysimilar
voters—say,Republicanmoms in their
40sinaMidwesternsuburb—andexpect
to complete at leastone interviewfrom
thatgroup of five. I’dbuild asample of
500 differentclusters of five votersper
cluster,or2,500 voters total. From that
number,Icould be reasonably assured
that500 peoplewould talk to us.The
500clustersweredesignedtorepresenta
diversecross-sectionoftheelectorate.
Astheyearsdriftedby,ittookmoreand
more voters per cluster for us to geta
single voter to agreetoaninterview.
Between 1984 and 1989, whencaller ID
was rolled out, more voters began to
ignoreourcalls.Theadventofanswering
machines and then voicemailfurtherre-
ducedresponses.Votersscreentheircalls
more aggressively,socooperation with
pollsters has steadily declined year by
year.Whereas once Icould extractone
complete interviewfromfive voters, it
cannowtakecallstoasmanyas100voters
tocompleteasingleinterview, evenmore
insomesegmentsoftheelectorate.
Andhere’sthekillerdetail:Thatsingle
cooperative soulwho speakswithan
interviewer cannotpossiblyholdthe
sameopinionsasthe99othervoterswho
refused.
In short, we no longer have trulyran-
dom samplesthatsupportclaims that
pollresultsaccuratelyrepresentopinions
oftheelectorate.
Instead, we have samples of “the will-
ing,” whatresearcherscalla“convenience
sample” of thoseconsenting to give us
their time and opinions.Despiteknowl-
edgeofthis, pollsters (including myself)
haveglossedoverthisrealitybydressing
upourresultswithclaimsofpollshaving
a“marginof error”ofthree or four per-
centage points when we knew, or should
haveknown,thattheerrorfactorisincal-
culable given the non-randomsample.
Mostpollstersturnedtoweightingresults
to“fix”variationsincooperation,butthis
can inadvertently amplifysampling er-
rorsduetononcooperation.
Forawhile,mostpollsconductedmost
ofthetimeinmostplacesseemedreason-
ably accurate, so we kept at it,claiming
randomsamplesurveyswithlowmargins
of error.Weighting becameaBand-Aid
for noncooperation. And polling still
seemed better than hoisting awetfinger
tothepoliticalwinds.Thencamethepast
twopresidential elections, exposing
deeperwounds.
IoffermyownexperiencefromFlorida
inthe2020electiontoillustratetheprob-
lem. Iconducted tracking polls in the

The dirty


little secret


pollsters need


to own up to


PHOTOBY SKODONNELL/GETTYIMAGES/ISTOCKPHOTO

o


Votersscreentheircallsmore


aggressively,so cooperation


with pollstershas steadily


declinedyear by year.

Free download pdf