Registration Of Voters.
For the first multiparty elections in 1992 the ECK conducted a national registration drive.
The data of registrants was entered in the so-called “black books” from which mimeographed lists for use at the polling stations were derived.
In 1997 the ECK computerised the registers using optical mark recognition (OMR) forms, although the black books were kept as a back-up.
The 2002 elections were based on the 1997 register, updated in registration drives in 2000, 2001 and 2002.
The use of black books was forbidden in those elections. (The same approach was to apply in the 2007 elections but shortly before the elections it was decided that black books could be used as back-up.)
There were two registration drives in 2007, from 1 to 30 March, reinforced by a second drive from 11 June to 31 July (with 30-day inspection periods following each drive).
The unusually long period of registration (67 days) produced good results. The number of registered voters increased by 1,767,212, resulting in a total number of voters for the 2007 election of 14,296,180. This represented 71% of the 19.8 million persons over 18 years of age who had been issued national ID cards.
The introduction of continuous registration, perceived as a forward step in 2002, has in practice created a system that combines the cost of both the periodic and the continuous systems of registration.
But the ECK still conducts yearly registration drives that in magnitude and cost are similar to those conducted in the case of a periodic register.
The system as it exists today is open to serious criticism: Continuous registration has not worked – only (2-3)% of the registration took place at the ECK offices.
The ECK alleges that this is because the number of field offices is too small (and aims to have an office in each constituency).
This is not correct: a significant proportion of the Kenyan population lives within a reasonable distance of an ECK district office . . . yet only a minimal fraction of that part of the population opts to register at ECK field offices.
Furthermore, the deletion of names of deceased voters from the register is ineffective: the Central Bureau of Statistics estimates that 1,733,000 persons have died since 1997 but the ECK has been able to eliminate the names of only 513,000 deceased persons from the register.
Statistically, therefore, the names of some 1.2 million dead persons swell the voter register.