20 Basic Statistical Methods for

Quality Control Workers

E. Wajnberg

INRA, 37, Blvd du Cap, 06600 Antibes, France

Introduction

Entomologists and/or producers of benefi-
cial insects and mites, who have to use qual-
ity control (QC) procedures, always have to
handle a lot of numerical data collected dur-
ing mass rearing. Once these data are col-
lected, tools are needed: (i) to summarize the
information available; and (ii) to check
whether their mass-reared animals satisfy
predefined criteria, such as size and fecun-
dity. The statistical methods that are used
can lead to wrong conclusions about the
quality of the natural enemies if the basic ter-
minology, assumptions and procedures are
not known and accurately understood. The
aim of this chapter is to provide the reader

with a minimum of knowledge needed to

summarize the data collected and to test

whether the mass-produced organisms meet

the preset quality criteria. In the first two

sections, the necessary terminology is speci-

fied and a short description of the usual sta-

tistical distributions that QC workers have to

handle is given. Normally, several traits are

quantified simultaneously on mass-reared

animals. Therefore, in the next sections, a

thorough description is given on how to

analyse each quantified trait separately,

together with a presentation of methods to

take two quantified traits simultaneously

into account.

The aim of this chapter is not to provide the

reader with a deep knowledge of statistical

© CAB International 2003. Quality Control and Production of Biological Control Agents:
Theory and Testing Procedures (ed. J.C. van Lenteren) 305

Abstract

The quality of mass-produced beneficial insects and mites needs to be checked regularly, but mass-rear-
ing companies often do not have personnel with knowledge of the statistical techniques that have to be
used to do this. Therefore, a number of simple and easy-to-perform statistical methods are described in
this chapter. Quality control workers normally collect a lot of numerical data, and these data should then
be handled using proper and accurate statistical methods, in order: (i) to describe the main features of the
mass-produced animals; and (ii) to check whether these animals satisfy the predefined standards. This
chapter provides the basic terminology and knowledge concerning the statistical methods for quality
control of mass-produced beneficials. Working examples are presented throughout the chapter, to help in
the understanding of the concepts of the use of statistical methods in this case. This chapter should be
considered as a starting-point, leading the interested reader to more advanced statistical methods, which
can be found in several textbooks that are mentioned at the end of the chapter.