![]() |
Inferenceveryard projects > knowledge management > inference |
on this page |
on this website
|
||||||||
How to help or hinder people from drawing conclusions |
|
Observations of Inference in Practice | Truth |
![]() |
Observationsveryard projects > knowledge management > inference > observations |
![]() |
People are liable to use single homogeneous sources of information - especially when they are unmotivated or ignorant of the available sources. |
![]() |
People are capable of using multiple and diverse sources of information
to support inference - when they are sufficiently motivated.
For example, checking independent websites, asking friends and family for recommendations, paying for specialist advice. |
![]() |
When the same information comes from multiple diverse sources, people
typically regard this as adding credibility. Confidence in the information
is increased.
For example, lots of journalists making similar evaluations of some commercial product or issue. |
![]() |
When inconsistent information comes from different sources (=interference),
people typically regard thus as undermining specific credibility. Confidence
in each information item is reduced. However, confidence in the general
mass of information may sometimes be increased.
For example, if there are many conflicting rumours about the sexual practices of certain celebrities, this may cause some people to dismiss all such rumours as rubbish, while others will fall for the "No Smoke Without Fire" fallacy. |
![]() |
People are generally not equipped to detect well-orchestrated campaigns of spin and propaganda, and are vulnerable to a range of inference tricks and errors. |
![]() |
People are often tolerant of imperfect information, and can be tricked into confirming the intelligent guesses of clever investigators. People underestimate the infererence power of a determined and unscrupulous investigator with access (legitimate or otherwise) to large quantities of data. |
![]() |
However, once people are alerted to particular sources of spin and propanda, they then systematically distrust any future information that can be easily identified as coming from these sources. |
![]() |
veryard projects > knowledge management > inference |
Copyright © 2003 Veryard Projects Ltd http://www.veryard.com/kmoi/inference.htm |