Tuesday, September 27, 2016

QUICK HITTERS- STATISTICS AND POLLS; PAY NO ATTENTION. PT 1.



Checklist, Lists, Business, Form, Checkbox, Document


Perhaps the two most misleading and least understood campaign tactics, used by candidates to win an election, are numbers reported through surveys, and conclusions drawn from statistical analysis of previous outcomes regarding certain subjects.

Both of these are easy to manipulate
as the following examples indicate;

WORD THE SURVEY QUESTION IN SUCH A WAY TO ENCOURAGE A CERTAIN RESPONSE;

Example- "Given the poor state of the economy, do you think Incumbent Candidate X can be trusted to fix the problem?"

See the real problems(s)?

#1-  The question does not in any way show justification for concluding that the economy is in "Poor" condition.

2) That there is a question of the "Trust Worthiness"  of Candidate X, yet there is no evidence given that there has been dishonesty in the past.

It is clear, that in cases such as this, that the wording of the question presumes a negative view of Candidate X. Even answering the question could be construed as agreeing with its assertions.  

MANIPULATING THE RESULTS IS TO EASY.

This should be a given, but it is often not reported or ignored. That Survey or Poll results can be greatly influenced by the Tactics of those conducting them.

Example-  How often do you here a poll prefaced by the words like...

"... among registered..."

"...survey of likely voters..."

"...who identify themselves as.."

"...popularity among age groups..."

What do they all have in common? They tell us virtually nothing.

-  Where did the questioning take place?  Particularly, the "COMMUNITY."  Even in a small city or town, political differences can vary greatly, so knowing the exact locations can give us an indication about the validity of reported results.

-  How much of the information acquired about those who were interviewed or polled, was self- reported or unverifiable?

-  What were the guidelines or standards set that could disqualify an individuals answers from being included in the final results?

-  If there was more than a single question asked, was each individual asked ALL of them, or did previous answers influence the length of the interview?

-  What protocols were in place to guarantee that the Interviewers or Pollsters followed proper procedures?
Look for part 2.