In the immediate aftermath of the 2007 Kenya election I exchanged emails with Joel Barkan who had just returned to Washington from the IRI/USAID Election Observation Mission. On January 2, 2008 Joel was trying to understand why the exit poll had not been released “to calm Raila’s people and perhaps prevent tomorrow’s march.” He wrote:“You know, if this is not released before six months out, both IRI and probably the embassy will be accused of a cover up.  I would reflect again on how this should be played . . .“.  This is my response:

Joel,

My e-mail shows that I responded to this message, but the “sent” box does not reflect this.  There was a connection problem and my copy of the text will not come up.  What I drafted was long and I will attempt to reconstruct in some fashion:

I completely agree with your thoughts.  At the urging of our polling firm and UCSD I argued this as vigorously as I knew how within IRI to no avail.  I see a major embarrassment in the works as time passes.

This was a much better poll than our previous exit polls in 2002 and 2005 in which we had expressed pride thanks to the tireless work of James Long/UCSD.  The previous sampling was 3,000 in 55 constituencies. {Ed. note: 2007 sample size was 5500 in 179 constituencies in 71 districts out of a total of 212 constituencies in 72 districts}

My original agreement with IRI DC was that we should not release data to anyone while the polls were still open even though USAID had said that they would like preliminary data that Strategic said would be available around 3pm.  This was consistent with agreed US practice to avoid influence on voting.  In spite of this, Sheryl pressed me while I was still at polling place on the afternoon/evening of the election saying that the primary reason for funding the poll was for “early intelligence”.  She got the data by calling Strategic directly.

I sent Sheryl an e-mail confirming that she had gotten the data from Strategic and that I understood that the data was for “internal use of USAID only” and not to go to anyone else.

Frankly, I was concerned that the data would get to the Kibaki camp and that they would make tactical use of it.

Background:  On Thursday, two week before the election, I got a message from Sheryl that the Ambassador needed a copy of our last delegate list asap and she sent me a fax for him.  I sent the list, noting that it was to be released to media the next day.  On my way to lunch I got a call on my cell from the Ambassador raising hell about Mark Bellamy being on the list, saying that he was “laying down a marker” and that he would hold me “personally responsible” as IRI’s “person on the ground” even though I had in previous conversations explained that I had little or no influence over the delegate selection in DC.

More to follow:  Just got an e-mail from our press office with a mention of our failure to release the poll in Slate.

Ken

Here is the link to that Slate piece “What’s going on in Kenya? And why didn’t a U.S.-funded group release its exit poll data?”.