INFORMATION ABOUT SWITCH PLATE TASKING/REPORTING/EVALUATION

• Document Type: CREST
• Collection: STARGATE
• Document Number (FOIA) /ESDN (CREST): CIA-RDP96-00789R002100270011-8
• Release Decision: RIFPUB
• Original Classification: S
• Document Page Count: 3
• Document Creation Date: November 4, 2016
• Document Release Date: November 1, 2002
• Sequence Number: 11
• Content Type: REPORT

Approved For Release 2002/11/18: CIA-RDP96-007898002100270011-8
S E C R ~: T STIPPLED NQFORN
SPECIAL ACCESS REQUIRED
INFORMATION ABOUT SWITCH PLATE TASKING/REPORTING/EVALUATION
The information provided by SWITCH PLATE sources is obtained through a
unique and highly sensitive collection technique. Your care in evaluating
this information will insure that we are better able to assess and/or modify
the technology to provide you, the customer, with a better product.
Information gathered utilizing this technique is likely to consist of a
mixture of correct, incorrect, and irrelevant or apparently irrelevant data.
You should be aware of the following facts:
1. Experience has shown that the Source's raw data, unedited, tends
to be of more value to the customer's Analyst, since it is more accurate
than someone else's summary of what the Source has said. The Project
Officer, however, may add his/her comments to the report as a separate
entry.
2. A more experienced Source will usually provide raw data only,
while a less experienced Source may be inclined to "pre-analyze" the
information when reporting it. While it is the Project Officer's
responsibility to assist in preventing the Source from doing this,
operational necessities such as criticality of reporting speed, violation of
item 1., above, or other constraints may allow some "pre-analyzed"
information to be forwarded.
EXAt1PLE: Amore experienced source ma,y describe the target (a submarine
under construction? as a huge, cylindrical, metallic object,
pointed on one end. A less experienced source may report the
object as "...It looks like some kind of a rocket".
An analyst familiar with this new technique and recognizing that ti~is
ha.s happened, will "back analyze", to determine what raw data caused the
Source to report such seemingly inaccurate data. To aid in a more complete
understanding of the Source's short answers, a brief summary, with sketches,
will be attached to the reporting sheet.
3. Sources, like all humans, tend to be attracted to aspects of the
target which attract them, personally. They also tend to ignore or gloss
aver aspects which do not attract them. This sometimes leads to widely
varying descriptions of the same target site. An old illustration serves as
an example:
5WP-HDA-0002-90
S E C R E T STIPPLED NOFORN
SPECIAL ACCESS REQUIRED
Handle Via SKEET Channels Only
COPY # L~
Approved For Release 2002/11/18: CIA-RDP96-007898002100270011-8
 Approved For Release 2002/11/18: CIA-RDP96-007898002100270011-8
S E C R E T STIPFI.1~lj NQFQRN
SPECIAL ACCESS REQUIRED
EXAMPLE: Four. blind men were tasked to report the shape of an elephant.
One stood in front, felt the trunk, and said that an elephant is
actually a huge variety of snake. Another felt the ear and
described an elephant as being like a living leaf, another felt
the tail, and reported that it is like a long, hairy rope. The
fourth felt one leg and reported that an elephant is a tall,
vertical animal, shaped like a tree trunk.
4. Sources may, at times, correctly answer a question in the wrong
terms. This is usually due to the Source's "pre-analyzing".
EXAMPLE: Returning to the Source who incorrectly labeled a submarine under
construction as a rocket: The Source might then report that it
will "blast off for the first time on May 10". While the identity
of the object is incorrect, the date given can be understood by
the Analyst as a prediction of the submarine's maiden voyage.
5. The Customer must always be mindful that Sources using this
technology tend to answer questions very literally. Vagueness in questions
will cause vagueness in answers. Questions/tasking must therefore be
carefully formulated. The Customer must understand that the more precise
he/she words a question, the more exact the Source's answer will be. The
rule of thumb is to ask a question which, to the best of the Customer's
knowledge, will have only a single, exact answer. Several examples may be
used to illustrate this:
EXAMPLE: Time spans should be as exact as possible? "Where will Mr. X be
next Thursday at 3PM?" is infinitely preferable to, "Where will
Mr. X be next week?" The Customer should avoid questions
containing phrases such as, "On ar about...". A worst possible
example of this would be, ""Where will Mr. X be within an
actionable period of time?'" If the Customer wants to know where
Mr. X will be throughout the span of next week, it would be
infinitely better to ask many questions of exact time: "Where will
X be at 3PM Monday?", "Where will X be at 3PM Tuesday?", etc.
EXAMPLE: Locations should be as exact as possible? "Describe the den of
X's home" is, on the surface, an exact question requiring an exact
answer. If, however, the Customer knows that X has several homes
and continually moves between them, it would be better to task the
Source with, "Describe the den in X's Los Angeles home."
EXAMPLE: The best results are obtained when onl one answer is ex ected to
SECRET STIPPLED NOFORN
SPECIAL ACCESS REQUIRED
Handle Via SKEET Channels Only
Approved For Release 2002/11/18: CIA-RDP96-007898002100270011-8
 Approved For Release 2002/11/18: CIA-RDP96-007898002100270011-8
S E C R E T STIPPLED NOFORN
SPECIAL ACCESS REQUIRED
each question: It is better to task the Source with "Will X be at
location A on Monday?", "Will X be at location B on Monday?", etc.
than to ask "Which of 5 known locations will X be at on Monday?"
(X may, in reality, visit several or none of them on Monday.)
Again, the rule of thumb is: Ask a question which, to the best of our
knowledge, will have only a single, exact answer.
6. The Customer should be aware that, for reasons of security and
operational effectiveness, Sources are not always given the background data
which the Customer provides with the tasking questions. Therefore, Sources
may tend to report extraneous data in their Summary Reports, which are
attached to the main Reporting sheet, The Summary Report usually contains
much irrelevant information, but may also contain choice tidbits of.untasked
information of possible value to the Customer's analysts, and may stimulate
further tasking.
A separate reporting package is returned to the Customer from each
Source. This package will include the Source's reporting sheet, the
Source's Summary page, and the Project Officer's comments.
As the receiving Analyst reads the answers and uses the information in
whatever way necessary, the EVALUATION WORKSHEET should be kept at hand. As
each answer is read from each Source, the Analyst can mark the evaluation
sheet to show the degree of usefulness that answer has been to him/her. The
evaluation criteria have been kept very simple:
VERY MARGINAL
USEFUL USEFUL
^ U
NO CANNOT BE
VALUE DETERMINED
When all Source Reports have been received and evaluated, please place
a xerox copy of the evaluation into the self-addressed stamped envelope
provided with it, and mail it to ACSI.
S E C R E T STIPPLED NOFORN
SPECIAL ACCESS REQUIRED
Handle Via SKEET Channels Only
Approved For Release 2002/11/18: CIA-RDP96-007898002100270011-8