Editor’s note: an attorney involved in the Capistrano Unified School District recall of trustees responded to an editorial in The Buzz last week, explaining how the Registrar of Voters performs a random sampling.

The Random Sample Process. In short, the Registrar will sample 5 percent of the total petitions turned in. They then figure out what percentage of those are valid (i.e., from registered voters in the district). For example, if 25,352 total signatures were turned in to recall one of the trustees, the Registrar would examine 1,268 (5 percent) of those to determine what percentage was valid.

 Assume the sampling shows 75 percent of the signatures were valid. The Registrar would then make an assumption that 75 percent of all the signatures turned in were valid. They would extrapolate that percentage onto the entire number of petitions turned in (75 percent of 25,352 = 19,014 valid signatures). 

Keep in mind, it takes 20,421 valid signatures per trustee for the Registrar to certify the recall -- NO MATTER WHAT. 

(a)    110 Percent for automatic approval. If the random sample process shows that more than 110 percent (22,463 = 110 percent of 20,421) of the total number required was obtained, then the Registrar will certify the recall without actually counting all the other signatures.

(b)     90 percent for automatic denial. If the random sample process shows that less than 90 percent (18,378 = 90 percent of 20,421) of the total number required was obtained, then the Registrar will deny the recall without actually counting all the other signatures.

(c)    Full count required if between 90 percent and 110 percent. If the random sample process shows that more than 90 percent (18,378) but less than 110 percent (22,463) of the total number required (20,421) was obtained, then the Registrar cannot rely upon the random sampling process and must count every one of the signatures turned in (25,352). 

Therefore, in my example, 19,014 is 93 percent of the required 20,421. As a result, the Registrar could not certify or deny the recall based only upon the random sample. That is why they were forced to count all of the signatures.  However, when they counted every signature they found, once again, that less than 20,421 valid signatures had been turned in. As a result, they denied certification.

It is possible that the Registrar's findings could be challenged and overturned if enough signatures were improperly invalidated.