FLOOD09697

• County: Statewide
• Community: Nationwide
• Basin: Statewide
• Title: Mathematical Modeling of a Sociological and Hydrologic Decision System
• Date: 6/1/1978
• Prepared By: Institute for Social Science Research on Natural Resources, Utah State Univ.
• Floodplain - Doc Type: Educational/Technical/Reference Information

from the standardized fonn of the Equation 5.17. It
is much larger than any other found in this form be-
cause of the obliteraling effect of an important single
summarizing variable as discussed under the section
on Equation 1Il.

All six terms in Equatinn 1Il were retained for
the regression run creating this equation so thai the
coefficients in Equation N could be compared with
the same terms in Equation 1Il under conditions where
every term is the same except for the new one added
(tenn 8 in Equation 5.16). Terms #2, # 3, and #6
of this same equation could have been removed since
their removal only reduced the r2 from .823 to .819.
However, even though these three tenns have liltle
effect in this particular application, they might .
have significant impacts on another situation} and
therefore, should be retained for future analyses. With-
out terms # 2, # 3, and # 6, and unstandardized fonn
of the equation is:
(I) (4) (5)
y =. + .587 - .178XI + .OOllX3Xll + .I02X4X12

(8)
+.442X7XI5 . (5.18)

The coefficients of Equation 5.18 have similar values
to those in Equation 5.6. Moreover, all of the coeffici-
ents in Equalion 5.18 are significant beyond the .05
level.

Equation IV (5.16) was placed in the model to
account for feedback from an implementalion agency
to a decision agency. In order 10 do this, there needs
to be an evaluation of a proposal in question by the
other agency with the previous evaluation of the de-
cision agency in mind. There also then needs to be a
re-evaluation by the decision agency with the input
from the other agency included. The two applications
of Equation IV in the model accomplish this.

Other agency i?Ilaluation (IMEV)

Inpu t from other agencies can influence the
flood control decision. Equation N (5.16) is used
in the model to represent this influence. The result-
ing recommendations is passed along to the decision
agency for a reconsideration of its initial evaluation
(Figure 5.1). If the plan isrejected at this point of
re-evaluation, this information together with the basis
of the rejection is transmilted back to the planning
agency.18 Equation N (5.16) is "calibrated" for a

18Thisrepresentation could be incorrect under certain
circumstances. The output of the other agency goes to the deci~
sion agency for consideration in it s revised evaluation of the
flood control plan. As the model is now constructed, this oc-
curs only when the output of the agency evaluation is positive.
However, it is conceivable that an objection could be ignored.
In this case, the response from the agency would still have an
effect which would be accounted for by the Type IV term in
Equation IV.

particular agency by the application of appropriale
specific values which have been delermined for rele-
vant variables within Equation IV (5.16). For exam-
ple, for a plaIIDing agency, the mean values for "Plan-
ning Agency" (Table 5.4) are substituted into Equa-
tion 5.16 to yieldl9.

y = + .491- .206Xl - .0102X9 + 3.24/XIO

+ .00469Xl1 + .0675X12 - .01l8X13 + .769X15

. (5.19)

Revised decision agency evaluation
(DARE)

liThe revised decision agency evaluation" is the
second of three evaluations by this agency as a pro-
posal moves through the system (Figure 5.1). Iflhe
decision resulting from this second evaluation is posi-
tive, the plan is announced to the general public. In
the model, the revised decision agency evaluation also
is given by an application of Equation IV, and the pre-
vious decision agency evaluations are input to term
#8 (Equation 5.16)in this application. The previous
decision agency evaluations have a significant effect
on the revised agency evaluation through interaction
with the other agency evaluation. This would be
characteristic of real life since a person or agency us-
ually tends toward consistency of viewpoint, especial-
ly in an area about which the group or person is famil-
lar and has taken a public stand. Inserting the decision
agency values from Table 5.4 in the unstandardized
form of Equation IV (5.16) gives the following equa-
tion which is an approximation of that which would
have been obtained had this equation been directly
calibrated to the decision agency.

y = +.491 - .206Xr .0190~ + 6.01/XIO

+ ,00359Xl1 + .216X12 - .0307X13 + .884X15

. . (5.20)

Equation VI: Final agency
i?Ilaluation (DAFE)

The regression on the data for this evaluation
was based on the same data as before but with terms
#7 and #9 added. The results are:

Unstandardized form:

(1) (2) (3)
y -.0466 - .210XI + .00522X2X9 + .419X2(l/XIO)
(4) (5) (6)
+ .00213X3Xl1 + .0988X4X12 - .0699X5X13

19S ct- -
ee IscuSSlon under Agency Evaluation Equations.

70