Carina Andersson’s research while affiliated with Lund University and other places

In a case study, the defect detection for functional test teams is investigated. In the study it is shown that the test teams not only discover defects in the features under test that they are responsible for, but also defects in interacting components, belonging to other test teams' features. The paper presents the metrics collected and the results as such from the study, which gives insights into a complex development environment and highlights the need for coordination between test teams in function test.

To contribute to the body of empirical research on fault distributions during development of complex software systems, a replication of a study of Fenton and Ohlsson is conducted. The hypotheses from the original study are investigated using data taken from an environment that differs in terms of system size, project duration, and programming language. We have investigated four sets of hypotheses on data from three successive telecommunications projects: 1) the Pareto principle, that is, a small number of modules contain a majority of the faults (in the replication, the Pareto principle is confirmed), 2) fault persistence between test phases (a high fault incidence in function testing is shown to imply the same in system testing, as well as prerelease versus postrelease fault incidence), 3) the relation between number of faults and lines of code (the size relation from the original study could be neither confirmed nor disproved in the replication), and 4) fault density similarities across test phases and projects (in the replication study, fault densities are confirmed to be similar across projects). Through this replication study, we have contributed to what is known on fault distributions, which seem to be stable across environments.

This article presents a spiral process model for an iterative case study on quality monitoring, conducted in an industrial environment. In a highly iterative project, everything seems to happen at the same time: analysis, design and testing. We propose a spiral process model for case studies, and present a study conducted according to the proposed process. In the study, metrics collected from three software development projects are analysed to investigate which characteristics are stable across projects and feature groups of the product. The contribution of the article is multi-fold, detailing the case study methodology used with its sub-goals and procedures. Furthermore, the article presents the metrics collected and the results as such from the case study, which gives insights into a complex development environment and the trends of the retrieved data. The analysed data serve as feedback to the project staff to facilitate identification of software process improvement. The data have also been used for defect prediction. Copyright © 2006 John Wiley & Sons, Ltd.

Replications are commonly considered to be important contributions to investigate the generality of empirical studies. By replicating an original study it may be shown that the results are either valid or invalid in another context, outside the specific environment in which the original study was launched. The results of the replicated study show how much confidence we could possibly have in the original study. We present a replication of a method for selecting software reliability growth models to decide whether to stop testing and release software. We applied the selection method in an empirical study, conducted in a different development environment than the original study. The results of the replication study show that with the changed values of stability and curve fit, the selection method works well on the empirical system test data available, i.e., the method was applicable in an environment that was different from the original one. The application of the SRGMs to failures during functional testing resulted in predictions with low relative error, thus providing a useful approach in giving good estimates of the total number of failures to expect during functional testing.

RESUMEN RESUMEN Serious efforts are spent on defect detection during software development. So, it's important to use the most efficient and effective defect detection methods . This guide can help practitioners choose efficient and effective defect detection methods using the evidence - based software engineering (EBSE) approach . Using 12 empirical studies , including 10 experiments and two case studies , the authors compare inspection and testing techniques and then derive practical implications . While research results are often inconclusive , the authors recommend that practitioners generally use inspections for requirements and design , and testing for code . Also , because different defect detection methods find different types of defects , the methods might be complementary . Finally, they list factors that help frame the question and guide practitioners in further steps to integrate and evaluate the detection methods in their environments.

Software inspections have been introduced in software engineering in order to detect faults before testing is performed. Reading techniques provide reviewers in software inspections with guidelines on how they should check the documents under inspection. Several reading techniques with different purposes have been introduced and empirically evaluated. In this paper, we describe a reading technique with the special aim to detect faults that are severe from a user’s point of view. The reading technique is named usage-based reading (UBR) and it can be used to inspect all software artefacts. In the series of experiments, a high-level design document is used. The main focus of the paper is on the third experiment, which investigates the information needed for UBR in the individual preparation and the meeting of software inspections. Hence, the paper discusses (1) the series of three experiments of UBR, (2) the individual preparation of the third experiment, and (3) the meeting part of the third experiment. For each of these three parts, results are produced. The main results are (1) UBR is an efficient and effective reading technique that can be used for user-focused software inspections, (2) UBR is more efficient and effective if the information used for UBR is developed prior to, instead of during the individual preparation, and (3) the meeting affects the UBR inspection in terms of increased effectiveness and decreased efficiency. In summary, the empirical evidence shows that UBR is an efficient and effective reading technique to be used by software organizations that produce software for which the user perceived quality is important.

Software inspection is an effective method to detect faults in software artefacts. Several empirical studies have been performed on reading techniques, which are used in the individual preparation phase of software inspections. Besides new experiments, replications are needed to increase the body of knowledge in software inspections. This paper presents a replication of an experiment, which compares usage-based and checklist-based reading. The results of the original experiment show that reviewers applying usage-based reading are more efficient and effective in detecting the most critical faults from a user's point of view than reviewers using checklist-based reading. This paper presents the data of the replication together with the original experiment and compares the experiments. The main result of the replication is that it confirms the result of the original experiment. This replication strengthens the evidence that usage-based reading is an efficient reading technique.

In order to understand the state of test process practices in the software industry, we have conducted a qualitative survey, covering software development departments at 11 companies in Sweden of different sizes and application domains. The companies develop products in an evolutionary manner, which means either new versions are released regularly, or new product variants under new names are released. The survey was conducted through workshop and interview sessions, loosely guided by a questionnaire scheme. The main conclusions of the survey are that the documented development process is emphasized by larger organizations as a key asset, while smaller organizations tend to lean more on experienced people. Further, product volution is performed primarily as new product variants for embedded systems, and as new versions for packaged software. The development is structured using incremental development or a daily build approach; increments are used among more process-focused organizations, and daily build is more frequently utilized in less process-focused organizations. Test automation is performed using scripts for products with focus on functionality, and recorded data for products with focus on non-functional properties. Test automation is an issue which most organizations want to improve; handling the legacy parts of the product and related documentation presents a common problem in improvement efforts for product evolution. Copyright

The two most common strategies for verification and validation, inspection and testing, are in a controlled experiment evaluated in terms of their fault detection capabilities. These two techniques are in the previous work compared applied to code. In order to compare the efficiency and effectiveness of these techniques on a higher abstraction level than code, this experiment investigates inspection of design documents and testing of the corresponding program, to detect faults originating from the design document. Usage-based reading (UBR) and usage-based testing (UBT) were chosen for inspections and testing, respectively. These techniques provide similar aid to the reviewers as to the testers. The purpose of both fault detection techniques is to focus the inspection and testing from a user's viewpoint. The experiment was conducted with 51 Master's students in a two-factor blocked design; each student applied each technique once, each application on different versions of the same program. The two versions contained different sets of faults, including 13 and 14 faults, respectively. The general results from this study show that when the two groups of subjects are combined, the efficiency and effectiveness are significantly higher for usage-based reading and that testing tends to require more learning. Rework is not taken into account, thus the experiment indicates strong support for design inspection over testing.

Process understanding and improvements are essential in software industry in order to achieve cost effectiveness and short delivery times. One means of increasing proc- ess understanding and improvement is to utilize software process simulation. This paper describes how a template model was cre- ated in order to increase the knowledge of the code de- velopment and test processes for an industrial organiza- tion. The template model was created from an existing system dynamics model for the unit test phase. The paper shows how the template model can be adapted and ex- tended to fit a similar organization. The simulation model is applied for investigating the relationship between de- fect prevention in the development phase and defect de- tection in the various test phases. Data from a large con- tract-driven project were used in a case study to calibrate an adapted and extended model, which included code development and four test phases. Programmers and test- ers were involved in the design of the model. The results show that it is possible to use the intro- duced template model and to adapt and extend it to a specific organization. We can also conclude that it is im- portant to involve project members who contribute to the model building. The process understanding of the partici- pating project members is increased due to their involve- ment.