FLOOD04444

• County: Statewide
• Basin: Statewide
• Title: Flood Flow Frequency for Ungaged Waterways: a Literature Evaluation
• Date: 11/1/1977
• Prepared By: USDA
• Floodplain - Doc Type: Educational/Technical/Reference Information

(4) professional journals would better serve the profession if they would
require authors to consider the criteria of accuracy, reproducibility, and
practicality, and if they would adopt a consistent set of indices that reflect
these criteria.

INTRODUCTION

The Work Group recognized that a literature review and evaluation would be
an important aid in achieving their objectives. This publication summarizes
the literature evaluation that was performed. The literature evaluation was
designed specifically to provide the Work Group with information that would be
helpful in selecting procedures that are representative of those currently in
use.

The Work Group agreed that there are three criteria involved in the selec-
tion process: accuracy, reproducibility, and practicality. A selected proce-
dure should give the highest degree of accuracy possible within the limitation
of a practical procedure that achieves consistent results. It is not possible
to detail the appropriate level for these three criteria. To the, extent pos-
sible, standard measures of accuracy, i.e., standard errors of test results,
should be used as a criterion to judge between procedures. Reproducibility
can be enhanced by adopting procedures that require a minimum number of
subjective decisions. Practicality, as measured, for example, by the cost of
estimation and the time required to make an estimate, is an important criterion
in procedure selection and must be weighed against the available resources.
The selected. procedures need not be restricted to those capable of field com-
putation by individuals with limited experience and education.

CRITERIA FOR COMPARISON OF PROCEDURES

One objective of this study was to identify publications that described
the evaluation of models on the basis of accuracy, reproducibility, and prac-
ticality, which are the three criteria established by the Work Group for
comparing procedures. The following are definitions for these three criteria:

Accuracy:

a measure of the closeness of the predicated values
to the "true" value of the quantity being evaluated.
It considers both precision and bias.

Reproducibility:

a measure of the degree to which the predicted values
of the quantity being evaluated by a group of indi-
viduals agree with each other.

Practicality:

a measure of the effort and resources required to
obtain an estimate of the quantity being evaluated.

2