Boks and Mayeda's Bogus Success Numbers

Ed Boks relies on statistics from Merritt Clifton, publisher of Animal People, to prove he is doing a great job.

Marcia Mayeda should give Clifton a big fat kiss on the cheek because he makes even her abysmal statistics look good by making huge errors in his calculations based on not knowing many facts.

Merritt uses the (stupid) statistical marker of city shelter success the number of animals killed per 1,000 population, rather than Winograd’s much more reasonable and common sense measure of the kill and live save rate.


Nathan considers saving the lives of 90% of the animals impounded to approximate no-kill.

Using Merritt’s criteria, NYC, which kills as many animals as LA, is basically no kill because its population is more than double LA’s. Divide 17,000 killed in LA with a 4 million population, compared to 17,000 killed in NYC with a population of 9 million, gives NYC less than half LA's rate (2.0 vs 4.3). Follow?

Boks on the other hand, uses Merritt’s ranking of LAAS’ performance of 4.3 killed per thousand (about 17,000 thousand killed in 2007), which gives a performance about 3 times better than the national average of 12.5 animals killed per thousand human population.


Boks does not use the kill and live save numbers and rates to show how well he is doing because they show he is not doing well (43% kill rate vs. 16% in San Francisco for example).

The Merritt/Boks measure is ridiculous. For example, if LAAS impounded 17,000 animals in 2007 and KILLED all of them, LAAS would have a kill rate of 100%, and have Merritt's kill rate of 4.3 animals killed per 1,000, which is 3 times better than the national average.


That is, kill everything that moves and, and by Merritt's measure, you are still doing a fantastic job.

Clifton’s methods and measures are ridiculous. Boks uses them, because they work for him; Bickhart, Kramer, Blackman and the Mayor neither understand or care. They accept Boks’ "proof" and give him an award for being the “lowest no-kill city in the country” according to Villarraigosa, who is English challenged.

Clifton made even a bigger error when it comes to Mayeda’s County killing machine.

Merritt simplistically, and incorrectly, subtracted LA City and Long Beach’s population from LA County’s total population of 10,000,000, which would leave Mayeda's County shelter area having a population of 5.5 million.


With 43,000 animal killed, County would have a kill rate of 8.5/1,000, well below the national average of 12.5.

In fact, Clifton made a huge error based on incomplete information, and when this was pointed out to him, he refused to recalculate or revise his conclusions. Sue Freeman directly confronted Clifton, Boks and Bickhart about Clifton's mistakes, but no one cared
.

She asked Ed and Jim to remove Clifton’s numbers from the LAAS website because they were so wrong.

Clifton's mistake was that he did not take into account that there were 5 other shelter systems in the County that served the entire County area; he didn't even bother to find out even after Sue pointed out his error.


The other 5 shelter systems (other than LA City) serving the larger County area:

Pasadena Humane Society:Total population: 483,413.

Santa Monica Animal Control: (91,439; 2007).

SEACCA: The total population covered by SEAACA is 763,922.

Burbank Animal Shelter: (108,029; 2007).

Inland Valley Humane Society: Total population covered by Inland Valley Humane Society is 384,289.

San Gabriel Humane Society: Total population 245,000

Clifton uses 5.5 million population for Mayeda’s County operational area, giving a kill rate of 8.5 per thousand, when in fact, her operation only covers a population of 3.7 million, which puts the actual County kill rate at 12.5 animals per 1,000, 50% higher than Merritt reported. How can anyone trust statistics that are off by 50%?

When Sue Freeman pointed out Clifton's errors, Boks and Bickhart refused to make any changes or acknowledge that Merritt made mistakes and was a loser. Merritt's chief claim to fame is that he is the editor of Animal People with a readership of 30,000.

Clifton makes this kind of error throughout all of his calculations about animal populations, whether pets in Baltimore's pounds, or dogs in Calcutta. He does not grasp the concept of city/county/regional boundaries and lumps them all together and extracts whatever conclusions he wants.

I emailed Clifton, questioning his methodology and numbers. I did not offer any alternative methods or opinions. I expected, like all scientists and statisticians, he’d defend his methods by providing numbers and the theories he operated under. Here is Merritt’s response:

"I have no interest in your opinions.

"They are not supported by a significant record of accomplishment.

"That's the bottom line. What I do works, & has worked in many situations, involving many different species, around the world."


Merritt Clifton
Editor, ANIMAL PEOPLE
P.O. Box 960
Clinton, WA 98236

Telephone: 360-579-2505
Fax: 360-579-2575
E-mail: anmlpepl@whidbey.com

A friend of mine emailed me:

“What is it that Clifton does that works?”

My response was, “What he does is get a lot of people to accept his numbers because neither he nor they have any idea of what he is talking about.”

I have challenged Clifton again on his methodology. When I get a response, I’ll post it. Don't hold your breath for a rational response.

In the meantime, just know that nothing that comes out of Merritt’s or Boks’ mouths has any credibility whatsoever.


Oh, added info. Merritt did respond in the comments below.
.

10 comments:

  1. "I have no interest in your opinions?".

    Wow, that's rude.

    ReplyDelete
  2. Anonymous15 May, 2008

    Thus is Merritt Clifton's response:

    If & when you run an animal control agency, humane society, municipal planning department, wildlife agency, or grant-giving foundation of significant size, & want help in designing a situation-specific response to an animal population-related problem, I will lend a hand--as I have to assist many hundreds of others over the years, on every inhabited continent.

    Until then, others need my time (my e-mail box is full of inquiries), & you need to find something useful to do with yours.
    .

    ReplyDelete
  3. Anonymous15 May, 2008

    Wow...think that guy has something to hide?

    ReplyDelete
  4. Anonymous15 May, 2008

    That is correct. You (Muzika) have no useful or active role in the issue.

    ReplyDelete
  5. Anonymous15 May, 2008

    Muzika:

    You have never set foot on the ground and counted animals.

    Merritt:

    I have in fact done this in representative neighborhoods of Milwaukee, just two years ago, which is what led to the Milwaukee discussion; many parts of the Los Angeles area at different times, most recently three years ago; many hundreds of places around the world.

    In January 2007, demonstrating my methodology, I did a two-hour survey of one particular suburb of Ahmedabad, India, that was followed up by a three-day survey coordinated by Animal Help Ahmedabad.

    Animal Help found that I had discovered exactly 90% of the dogs, including all of the street dogs. The 10% I missed were basically the indoor lap dogs. With an appropriate adjustment of the compensation factor for possible unseen indoor animals, who leave no visible outdoor tracks or scat, and don't bark at passing strangers, future counts in that area should be bang on target.

    ReplyDelete
  6. Anonymous15 May, 2008

    So if understand correctly, 2 hours of calculations resulted in a
    number approximately equal to 90% of what was counted during a 3 day (door
    to door?) survey of some city in India with a human population of over 5 million?

    Interesting.

    -brad

    ReplyDelete
  7. Anonymous15 May, 2008

    Two hours of a foot survey to determine habitat occupancy.

    ReplyDelete
  8. Merritt, you conducted a two hour, on the groud survey in a City of 5,000,000 people and maybe 1 million households, and your 2 hour survey of "habitats" came up with almost the same number of dogs as a THREE DAY SURVEY, of unknown size or validity by Animal Help?

    No wonder you are able to pull statistics out of your hat. Neither your survey or Animal Help would be statistically significant, even if you both conducted your two surveys completely isolated and independent of the other.

    How many "habitats could you "survey" in two hours?

    How many did Animal Help survey in three days?

    When I was with the Census Bureau, we took five weeks to survey just the housed LA population and another two weeks to count the homeless, and we had over 5,000 census takers.

    A two hour survey and a 3 day survey?

    ReplyDelete
  9. Is this seriously Merritt Clifton?

    You do not have time to respond to email in a gracious manner, yet you have time to comment on people's blogs?

    I emailed an expert on calici virus in cats - http://www.cvm.ncsu.edu/docs/richard_ford.html

    and received a prompt, accurate response within 24 hours.

    I wonder if he has time to comment on blogs.

    ReplyDelete
  10. Wow, is this seriously Merritt Clifton?

    I had cause to email an expert in feline calici virus one day this week.

    http://www.cvm.ncsu.edu/docs/richard_ford.html

    I received a prompt, accurate and gracious response within 24 hours.

    I wonder if he has time to comment on people's weblogs.

    Particularly those people's weblogs in whose opinion he has no interest.

    Personally, I don't read people's weblogs in which I don't have interest. I would find that to be bizarre behavior.

    ReplyDelete