The Risk of Using the Q Heterogeneity Estimator for Software Engineering Experiments

All meta-analyses should include a heterogeneity analysis. Even so, it is not easy to decide whether a set of studies are homogeneous or heterogeneous because of the low statistical power of the statistics used (usually the Q test). Objective: Determine a set of rules enabling SE researchers to find out, based on the characteristics of the experiments to be aggregated, whether or not it is feasible to accurately detect heterogeneity. Method: Evaluate the statistical power of heterogeneity detection methods using a Monte Carlo simulation process. Results: The Q test is not powerful when the meta-analysis contains up to a total of about 200 experimental subjects and the effect size difference is less than 1. Conclusions: The Q test cannot be used as a decision-making criterion for meta-analysis in small sample settings like SE. Random effects models should be used instead of fixed effects models. Caution should be exercised when applying Q test-mediated decomposition into subgroups.

Quantitative Determination of the Relationship between Internal Validity and Bias in Software Engineering Experiments: Consequences for Systematic Literature Reviews

Quality assessment is one of the activities performed as part of systematic literature reviews. It is commonly accepted that a good quality experiment is bias free. Bias is considered to be related to internal validity (e.g., how adequately the experiment is planned, executed and analysed). Quality assessment is usually conducted using checklists and quality scales. It has not yet been proven;however, that quality is related to experimental bias. Aim: Identify whether there is a relationship between internal validity and bias in software engineering experiments. Method: We built a quality scale to determine the quality of the studies, which we applied to 28 experiments included in two systematic literature reviews. We proposed an objective indicator of experimental bias, which we applied to the same 28 experiments. Finally, we analysed the correlations between the quality scores and the proposed measure of bias. Results: We failed to find a relationship between the global quality score (resulting from the quality scale) and bias; however, we did identify interesting correlations between bias and some particular aspects of internal validity measured by the instrument. Conclusions: There is an empirically provable relationship between internal validity and bias. It is feasible to apply quality assessment in systematic literature reviews, subject to limits on the internal validity aspects for consideration.

Configuration management and product lines to enhance the replication process in software engineering

This research is concerned with the experimental software engineering area, specifically experiment replication. Replication has traditionally been viewed as a complex task in software engineering. This is possibly due to the present immaturity of the experimental paradigm applied to software development. Researchers usually use replication packages to replicate an experiment. However, replication packages are not the solution to all the information management problems that crop up when successive replications of an experiment accumulate. This research borrows ideas from the software configuration management and software product line paradigms to support the replication process. We believe that configuration management can help to manage and administer information from one replication to another: hypotheses, designs, data analysis, etc. The software product line paradigm can help to organize and manage any changes introduced into the experiment by each replication. We expect the union of the two paradigms in replication to improve the planning, design and execution of further replications and their alignment with existing replications. Additionally, this research work will contribute a web support environment for archiving information related to different experiment replications. Additionally, it will provide flexible enough information management support for running replications with different numbers and types of changes. Finally, it will afford massive storage of data from different replications. Experimenters working collaboratively on the same experiment must all have access to the different experiments.

Definition of a support infrastructure for replicating and aggregating families of software engineering experiments

Experimental software engineering includes several processes, the most representative being run experiments, run replications and synthesize the results of multiple replications. Of these processes, only the first is relatively well established in software engineering. Problems of information management and communication among researchers are one of the obstacles to progress in the replication and synthesis processes. Software engineering experimentation has expanded considerably over the last few years. This has brought with it the invention of experimental process support proposals. However, few of these proposals provide integral support, including replication and synthesis processes. Most of the proposals focus on experiment execution. This paper proposes an infrastructure providing integral support for the experimental research process, specializing in the replication and synthesis of a family of experiments. The research has been divided into stages or phases, whose transition milestones are marked by the attainment of their goals. Each goal exactly matches an artifact or product. Within each stage, we will adopt cycles of successive approximations (generateand- test cycles), where each approximation includes a diferent viewpoint or input. Each cycle will end with the product approval.

Replications of software engineering experiments

There are many open issues that must be addressed before the replication process can be successfully formalized in empirical software engineering research. We define replication as the deliberate repetition of the same empirical study for the purpose of determining whether the results of the first experiment can be reproduced. This definition would appear at first glance to be good. However, it needs several clarifications that have not yet been forthcoming in software engineering: – What is the exact meaning of the same empirical study? Namely how similar should an experiment be to the baseline study for it to be considered a replication? What is the exact meaning of a result being reproduced? Namely how similar does a result have to be to the result of the baseline study for it to be considered reproduced? These and other methodological questions need to be researched and tailored for empirical software engineering.

A Case Study of Agile Software Engineering Methods and Practices in the MeeGo Harmattan Open Source and Product Context

The aim of this master’s thesis is to study how Agile method (Scrum) and open source software are utilized to produce software for a flagship product in a complex production environment. The empirical case and the used artefacts are taken from the Nokia MeeGo N9 product program, and from the related software program, called as the Harmattan. The single research case is analysed by using a qualitative method. The Grounded Theory principles are utilized, first, to find out all the related concepts from artefacts. Second, these concepts are analysed, and finally categorized to a core category and six supported categories. The result is formulated as the operation of software practices conceivable in circumstances, where the accountable software development teams and related context accepts a open source software nature as a part of business vision and the whole organization supports the Agile methods.

A software engineering approach to distributed systems for composite modeling

The goal of this thesis is to define and validate a software engineering approach for the development of a distributed system for the modeling of composite materials, based on the analysis of various existing software development methods. We reviewed the main features of: (1) software engineering methodologies; (2) distributed system characteristics and their effect on software development; (3) composite materials modeling activities and the requirements for the software development. Using the design science as a research methodology, the distributed system for creating models of composite materials is created and evaluated. Empirical experiments which we conducted showed good convergence of modeled and real processes. During the study, we paid attention to the matter of complexity and importance of distributed system and a deep understanding of modern software engineering methods and tools.

Using configuration management and product line software paradigms to support the experimentation process in software engineering.

There is no empirical evidence whatsoever to support most of the beliefs on which software construction is based. We do not yet know the adequacy, limits, qualities, costs and risks of the technologies used to develop software. Experimentation helps to check and convert beliefs and opinions into facts. This research is concerned with the replication area. Replication is a key component for gathering empirical evidence on software development that can be used in industry to build better software more efficiently. Replication has not been an easy thing to do in software engineering (SE) because the experimental paradigm applied to software development is still immature. Nowadays, a replication is executed mostly using a traditional replication package. But traditional replication packages do not appear, for some reason, to have been as effective as expected for transferring information among researchers in SE experimentation. The trouble spot appears to be the replication setup, caused by version management problems with materials, instruments, documents, etc. This has proved to be an obstacle to obtaining enough details about the experiment to be able to reproduce it as exactly as possible. We address the problem of information exchange among experimenters by developing a schema to characterize replications. We will adapt configuration management and product line ideas to support the experimentation process. This will enable researchers to make systematic decisions based on explicit knowledge rather than assumptions about replications. This research will output a replication support web environment. This environment will not only archive but also manage experimental materials flexibly enough to allow both similar and differentiated replications with massive experimental data storage. The platform should be accessible to several research groups working together on the same families of experiments.

The Innovation Sources in Software Engineering: A Systematic Literature Review

During the last century many researches on the business, marketing and technology fields have developed the innovation research line and large amount of knowledge can be found in the literature. Currently, the importance of systematic and openness approaches to manage the available innovation sources is well established in many knowledge fields. Also in the software engineering sector, where the organizations need to absorb and to exploit as much innovative ideas as possible to get success in the current competitive environment. This Master Thesis presents an study related with the innovation sources in the software engineering eld. The main research goals of this work are the identication and the relevance assessment of the available innovation sources and the understanding of the trends on the innovation sources usage. Firstly, a general review of the literature have been conducted in order to define the research area and to identify research gaps. Secondly, the Systematic Literature Review (SLR) has been proposed as the research method in this work to report reliable conclusions collecting systematically quality evidences about the innovation sources in software engineering field. This contribution provides resources, built-on empirical studies included in the SLR, to support a systematic identication and an adequate exploitation of the innovation sources most suitable in the software engineering field. Several artefacts such as lists, taxonomies and relevance assessments of the innovation sources most suitable for software engineering have been built, and their usage trends in the last decades and their particularities on some countries and knowledge fields, especially on the software engineering, have been researched. This work can facilitate to researchers, managers and practitioners of innovative software organizations the systematization of critical activities on innovation processes like the identication and exploitation of the most suitable opportunities. Innovation researchers can use the results of this work to conduct research studies involving the innovation sources research area. Whereas, organization managers and software practitioners can use the provided outcomes in a systematic way to improve their innovation capability, increasing consequently the value creation in the processes that they run to provide products and services useful to their environment. In summary, this Master Thesis research the innovation sources in the software engineering field, providing useful resources to support an effective innovation sources management. Moreover, several aspects should be deeply study to increase the accuracy of the presented results and to obtain more resources built-on empirical knowledge. It can be supported by the INno- vation SOurces MAnagement (InSoMa) framework, which is introduced in this work in order to encourage openness and systematic approaches to identify and to exploit the innovation sources in the software engineering field.

Empirical study on the maintainability of Web applications: Model-driven Engineering vs Code-centric

Model-driven Engineering (MDE) approaches are often acknowledged to improve the maintainability of the resulting applications. However, there is a scarcity of empirical evidence that backs their claimed benefits and limitations with respect to code-centric approaches. The purpose of this paper is to compare the performance and satisfaction of junior software maintainers while executing maintainability tasks on Web applications with two different development approaches, one being OOH4RIA, a model-driven approach, and the other being a code-centric approach based on Visual Studio .NET and the Agile Unified Process. We have conducted a quasi-experiment with 27 graduated students from the University of Alicante. They were randomly divided into two groups, and each group was assigned to a different Web application on which they performed a set of maintainability tasks. The results show that maintaining Web applications with OOH4RIA clearly improves the performance of subjects. It also tips the satisfaction balance in favor of OOH4RIA, although not significantly. Model-driven development methods seem to improve both the developers’ objective performance and subjective opinions on ease of use of the method. This notwithstanding, further experimentation is needed to be able to generalize the results to different populations, methods, languages and tools, different domains and different application sizes.

Specification and Evaluation of Safety Properties in a Component-Based Software Engineering Process

Over the past years, component-based software engineering has become an established paradigm in the area of complex software intensive systems. However, many techniques for analyzing these systems for critical properties currently do not make use of the component orientation. In particular, safety analysis of component-based systems is an open field of research. In this chapter we investigate the problems arising and define a set of requirements that apply when adapting the analysis of safety properties to a component-based software engineering process. Based on these requirements some important component-oriented safety evaluation approaches are examined and compared.

Component-based software engineering: a quantitative approach

Dissertação apresentada para a obtenção do Grau de Doutor em Informática pela Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia