Statistical Evaluation of Packing
House Waste Data
CONRAD  P.   STRAUB
Senior Sanitary Engineer
Environmental Health Center, U. S. Public Health Service
Oak Ridge National Laboratory
Oak Ridge, Tennessee
Some years ago, in studying the published data relating to the characteristics of packing house and slaughterhouse wastes, the writer was
impressed with the variability of the reported values. In an attempt
to group the data from the many plants, and to obtain a more reliable
estimate of the normal range in values that might be expected in the
volume, biochemical oxygen demand (B.O.D.), suspended solids (SS),
total nitrogen, grease, and population equivalent (PE) of the wastes,
these data have been analyzed by statistical methods.
According to Peters and Van Voorhis (1), in giving an adequate
description of a mass of quantitative data, one must do one or another,
or several, of the following things: 1) indicate the general size of
the variables—mean, mode, median, or other index of "central tendency"; 2) indicate the spread of the variables; 3) show the shape
of the distribution; 4) show the relation of two or mote sets of variables
to each other; and 5)  indicate dependability of the generalizations.
Methods that may be employed in evaluating the above criteria
have been described by Rair (2) and Velz (3). In recent years,
several papers have been published in which similar procedures were
used in the analysis of specific waste materials. For example, Weston,
Merman, and DeMann (4) (5) utilixed these techniques in evaluating
the probability of occurrence of different characteristics of oil waste
waters, Dougherty (6) for defining the characteristics of food processing wastes, and Eckenfelder (7) for describing the characteristics of
tomato and apple wastes. In the above studies, statistical procedures
were used to evaluate the wastes from specific plant investigations.
With the exception of these studies, the data reported in the literature
are generally limited to a listing of the mean or range without any
statement of the statistical variation of the data.
The evaluation of packing house wastes data, which is to be reported,
includes data from many plants, and is believed to be the first extensive
statistical evaluation of this type for these waste materials.
222