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

The r2 equals .548 with alllerms significanl at the .05
level. Equation 5.29 was not used in the model be-
cause il is more difficult 10 interpret3S and because
Equation 5.28 had a higher r2.

The r2 of Equation 5.28 was greater than that
of Equation 5.27 because the categorization of terms
accounled for nonlinearity in the relationships.39
Equation 5.28 was the one used in the model to pre-
dict public response 10 a flood control proposal. Un-
like the agency evaluation equations, public response
could be slightly negative and still not cause rejection
of a proposal. The value set for rejection in the model
is -.5.

IF statement C] to Cs

The nexl step in the model is to check for spec-
ial interest opposition to the flood control proposal
with five IF statemenls. The slalements are set 10
mirror the ability of special inlerest groups to block
projects which adversely affecl aspects about which
they are concerned. They specify limits of accept-
ability of factors such as cost, effectiveness, and eco-
logy.

Acceplance functions are used in these checks
rather than the perceived characteristics because a neg.
ative response by a special interest group to a proposal
would meet more success if that factor was also con.
sidered importanl by the general population. The ac-
ceptance functions reflect this interactive interrelation.
ship. A negative value for a factor of a proposal in the
acceptance function can become more negative if
either the imporlance factor (IF) increases or a pro-
posal is considered worse because of the factor. If a
factor is llllimportant, a severe adverse effect would
be required to cause a large enough negative value of
the acceptance function for the flood control proposal
to be rejected. If the factor were perceived as very
important, even a small negative value on the perceived
project characteristic would block adoption.

For some factors, the minimum acceptable value
will be positive. This kind of factor would be one with
respect to which the flood control proposal must actu-

38Standardization of this type of equation is not very
meaningfulsince the factors are not in separate terms. A power
function equation is best interpreted by remembering that
the exponents are weightings of the logarithms of the vari.
abies in relation to the log of the dependent variable.

39 Another way of accounting for nonlinearity is by spec-
ifying the non-functional form more precisel}'; however, actual
ratio level data are required for this to be done with confi-
dence. Recently, this has become possible in sociology (foot-
note 31), and a few analyses have achieved r2 values prior to
those obtained in the applied physical sciences (see Hamblin,
1974). Stimulus-attitude relationships appear to ofte:n have
a power-function relationship as do physical-stimulus res-
ponses (Hamblin, 1974).

ally improve conditions in order to be accepted. For
flood control proposals, one such faclor would be the
ability of the proposal 10 control flooding.

The relationships used by the model in the C
IF functions are:

IF statement Cl for recreation:40
(8)
Reject If XSX14 .;; -3.0

. (5.30)

IF statement C2 for ecology:
(9)
Reject If~X15';;-5.0. .

. (5.31)

IF stalement C3 for aesthetics:
(7)
Reject If X7X13 .;; -1.0

. (5.32)

IF statement C4 for effectiveness:
(6)
Reject IfX6X12 ';;4.0

. (5.33)

IF statement Cs for cost:
(10)
Reject IfX10X16 ';;.20.0.

. (5.34)

The variables are those defined in Table 5.6. The
values of the limits should nol be compared with each
other as a common scale of measurement was not used.

The values of the acceptance functions for IF
statements C1 to Cs used as threshold acceptabilily
levels are established by system simulation. They are
adjusted to values just below those that historically
caused plans to be rejected and just above those for
proposals that were accepted. This is done by exam-
ining historical records for the reasons, especially in
the actions of special interest groups, and adjusting
the appropriate acceptance function threshold levels.

These IF funclions generally worked well. How-
ever, IF function Cs (5.34) gave unrealistic responses
to cost in some cases because of the way in which one
of the variables in the population acceptance function
was handled. In an attempt to achieve consistency in
this term, willingness to pay, PAYL, and the recipro-
cal of perceived cost were used together41 in Ihe popu-
lation evaluation equation. This means that the value
of Ihe cost acceptance function goes up with willing-
ness to pay if the perceived low cost is positive and
down if it is negative. An increase in P A YL should al-
ways make a proposal more acceptable. It would have
been better, in this case only, to have not subtracled

40The converse is used in the computer program for
Cl, C2. C3. and C4; accept when the acceptance function is
greater than the value. This is exactly equivalent.
41See discussion on Interaction Terms.

79