Page:CIAdeceptionMaximsFactFolklore 1980.pdf/26

C00036554

FIGURE 4: DECEPTION, CRY-WOLF SYNDROME AND SURPRISE: EMPIRICAL EVIDENCE IN SUPPORT OF FOLKLORE

A. RAW DATA

WAS WOLF CRIED?	WAS DECEPTION ATTEMPTED?	WAS SURPRISE ACHIEVED?		TOTALS OR SUBTOTALS
		Yes	No
Yes	Yes No	23 1	0 2	23 3
No	Yes No	106 26	9 57	115 63
TOTALS OR SUBTOTALS		156	68	224

$\Downarrow$

WAS DECEPTION ATTEMPTED?	WAS SURPRISE ACHIEVED?		TOTALS OR SUBTOTALS
	Yes	No
Yes	129 (93%)	9 (7%)	138
No	27 (31%)	59 (69%)	86
TOTALS OR SUBTOTALS	156	68	224

WAS WOLF CRIED?	WAS SURPRISE ACHIEVED?		TOTALS OR SUBTOTALS
	Yes	No
Yes	24 (97%)	2 (8%)	26
No	132 (67%)	66 (33%)	198
TOTALS OR SUBTOTALS	156	68	224

$\chi ^{2}=$		96.59 uncorrected	$\chi _{1}^{2}(0.05)=3.84$
		93.68 corrected for continuity

$\chi ^{2}=$		7.15 uncorrected	$\chi _{1}^{2}(0.05)=3.84$
		5.99 corrected for continuity

$\Downarrow$

C. PARTITIONS OF $\chi ^{2}$

TERM	DEGREES OF FREEDOM	COMPONENTS OF $\chi ^{2}$ (Parameters Estimates)	SIGNIFICANCE
First Degree
Deception	1	0	N/A
Cry Wolf	1	0	N/A
Surprise	1	0	N/A
Interaction
Deception X Cry Wolf	1	8.968	> 1%
Deception X Surprise	1	96.591	>1%
Cry Wolf X Surprise	1	7.147	>2%
Deception X Cry Wolf X Surprise	1	0.153	-
TOTALS	7	112.659

Method of Computation Reference: Kendall, M.C and Stuart, A., The Advanced Theory of Statistics, Vol.2, Mafner, New York (1961), p.580 et. seq.

$\Downarrow$

D. Conclusions

• Deception is strongly associated with surprise. When deception was employed, surprise resulted in 93% of the cases; whereas, when deception was not used, surprise resulted in only about one-third of the cases.
• "Cry-Wolf" is likewise associated with surprise, though differences are less dramatic.
• The data are consistent with, but fall short of provinq, the hypothesis that deception and prior desensitization lead to even greater chances of surprise.