OFFICE OF RESEARCH OPINION ANALYSIS
DEPARTMENT OF STATE • WASHINGTON, DC 20520
March 19, 2007 SENSITIVE BUT UNCLASSIFIED - M-37-07
Nigerian Voters Coalesce Regionally Around Buhari
Polling data from late January in key Nigerian states indicates that although most Nigerian newspapers
continue to refer to the ruling party candidate (PDP/ Yar’adua) as the front runner, the opposition ANPP/
Buhari ticket may be more popular. However, currently it looks doubtful that Buhari, whose support is
mainly in the North, can pick up enough votes in the South to meet the electoral requirements for victory.1
When asked what presidential party they would be voting for, 37 percent of Nigerians polled in key states
name Muhammadu Buhari’s All Nigeria Peoples Party (ANPP), while 29 percent name the governing
Peoples Democratic Party (PDP), and 15 percent Vice President Atiku Abubakar’s Action Congress (AC).
Support for each party splits along regional lines: three-quarters of those in Kano and a plurality in Abuja
intend to vote for the ANPP. Meanwhile, the south is a contest between the PDP and the AC: the PDP
currently holds the lead in Enugu and Rivers, whereas the AC is slightly more popular in Lagos.
Nigerians’ party preferences shape who they expect will win the presidency. Overall, nearly four-in-ten
predict Buhari to win, while three-in-ten predict a Yar’adua victory. Nigerians are less confident in
Atiku’s success, perhaps due to the numerous obstacles his campaign is facing; indeed, it is not clear that
his name will even be on the ballot. Regionally, views correspond to patterns of party support, with many
more in the North anticipating a victory by the ANPP’s Buhari.
[1Electoral law requires the winner to obtain the highest number of votes and win at least 25 percent of
the ballots cast in 24 states. Results are based upon findings from a face-to-face survey conducted in five
Nigerian states (Abuja, Kano, Lagos, Enugu and Rivers) between January 17th and January 29th 2007.
In theory, the margin of error at the national level is +/- 3 percent. For further details, please see the
on page 4. Report prepared by Cara J. Carter (202-203-7927;
Character traits showcase Buhari’s appeal.
Nigerian perceptions of candidates’ character traits may help explain support for candidates. Of all the
candidates, Buhari is most frequently associated with integrity, leadership, accountability, honesty and
transparency. Nigerians are more skeptical about these qualities in Atiku and Yar’adua.
Atiku supporters could determine outcome of election.
Nigerian election law requires the winner to obtain the highest number of votes and win at least 25% of
the ballots cast in two-thirds (24) of the thirty-six states. Currently, Buhari holds an overall lead, yet he
has little southern support. If Atiku is unable to get his name on the ballot, AC supporters, most of whom
are in the South, could help Buhari meet the electoral threshold required for victory.
Assuming Atiku isn’t on the ballot, an analysis comparing AC/Atiku supporters’ views of both Buhari and
Yar’adua sheds light on which candidate they may support. Those in the South view Yar’adua more
favorably on most aspects, while those in the North lean towards Buhari. Thus, while Buhari may pick
up some votes from Atiku supporters, the data suggests that those in the South, where he needs the most
support, tend to look more favorably upon Yar’adua.
Buhari’s campaign limited to North.
Buhari’s limited reach with his ANPP campaign is one additional indicator of his struggle to gain a
foothold in the South. Although 72 percent of Nigerians mention that they have come in contact with
campaign activities, those in the South indicate little exposure to the ANPP.
Registration, although delayed, was successful. Although there was considerable criticism of the
registration process, eight-in-ten report being registered to vote.
APPENDIX A. HOW THE SURVEY WAS CONDUCTED
Poll results are from a January 17- January 27, 2007 public opinion survey conducted by a reputable
Nigerian research firm. Face-to-face interviews were conducted with 2500 adult Nigerians (aged 18-60)
in five key states: Lagos, Enugu, Abuja, Kano, and Rivers. These were intended to be representative of
five of Nigeria’s six regions, with only the North East left unrepresented. Sixty percent of the interviews
were conducted in urban locations and forty percent were conducted in rural areas.
As a guideline for interpretation, in 19 cases out of 20, results based on random samples of this size will
differ by no more than about 3 percentage points in either direction from what would have been obtained
by polling all residents in the target areas. The margin of error for differences between regional and
demographic subgroups is larger. In addition to sampling error, the practical difficulties of conducting
a survey of public opinion may introduce other sources of error into the results.
For additional information on the methodology of the survey, or for further analysis of the data, please
contact Cara J. Carter (CarterCJ@state.gov, 202-203-7927).
SENSITIVE BUT UNCLASSIFIED