Hardware Reference
In-Depth Information
Factors with Negative Influence on Software Testing
Practice in Spain: A Survey
Luis Fernández-Sanz
1
, M. Teresa Villalba
2
, José Ramón Hilera
1
,
and Raquel Lacuesta
3
1
Depto. De C. de la Computación, Universidad de Alcalá, Ctra. Madrid-Barcelona Km 33,600,
Alcalá de Henares, 28871, Madrid, Spain
2
Depto. de Sistemas Informáticos, Universidad Europea de Madrid,
C/Tajo s/n, Villaviciosa de Odón, 28670, Madrid, Spain
3
Dept. of Comp. and Syst. Eng., Univ. de Zaragoza, C. Escolar s/n, 44003 Teruel, Spain
luis.fernandezs@uah.es, maite.villalba@uem.es,
jose.hilera@uah.es, lacuesta@unizar.es
Abstract.
Software testing is the commonest technique for software quality as-
surance. It is present in every development project and concentrates a large per-
centage of effort, there are still not many studies which address the real practice
of individuals and organizations. Anyway, practitioners usually agree with the
idea that software testing efficiency and effectiveness in their organizations
might be improved. Two previous studies in Spain have revealed implemented
testing practices in organizations and individual performance of software pro-
fessionals when designing test cases should be improved. This paper presents
the results of a survey designed to know if 23 factors determined by a panel of
experts in 2007 may explain this situation of testing practice. Data collected re-
veal that none of the factors is clearly rejected as a negative influence for test-
ing although some of them are not generally accepted. Exploratory statistical
analysis reveals relations between certain pairs of items as well as a new group-
ing in factors.
Keywords:
Software testing, survey, influence factors.
1 Introduction
Software testing is the commonest techniques for verification and validation in devel-
opment projects. Every project includes a specific phase for testing and debugging.
According to different statistical studies of effort distribution throughout the life cycle
[1][2][3], this phase usually requires around a large percentage, around one-third
(ranging from 30 to 35%), of the total effort of the project.
Different studies have tried to analyze real practice but in many cases empirical
works are focused on analyzing or demonstrating the benefits of specific methods or
approaches to testing. As stated in [4], there is a need of real practice empirically-based
data not vested by such purpose but aimed at providing more light on this area. This
type of studies is rare although, as can be seen in the following sections, there are inter-
esting contributions. In order to gain knowledge in this area, a series of studies centered
