Software maintainability is one of the most crucial quality attributes of a software product. Software engineering researchers and practitioners have devoted considerable effort to developing “good design” methods, rules and principles to improve software maintainability. But before we can validate the effectiveness of these methods, we first need an approach to measure software maintainability. The existing maintainability measures usually have limited scope and accuracy since they either isolate the software from its environment and focus only on the technical properties of the software, or measure a confounding effect of various factors involved in the maintenance process. Furthermore, these measures are often defined and collected on a coarse-grained level and provide no insight into what makes software difficult to change. This research addresses the problems associated with software maintainability measurement by adapting the concepts of task complexity from the human behavior domain to the software engineering domain. This dissertation involves developing and validating a measurement model for measuring the maintainability of software, to provide a better understanding of the difficulty in modifying software and the effect of software design methods on software maintainability. A measurement protocol and a tool have been developed to support the application of the measurement method.

This research introduces the concept of transferability into the usability construct and creates the Usability Paradigm for Multiple Device System (UPMDS) to conceptualize and quantify the usability in multiple device scenarios. This study fills the literature gap that no effective method exists in measuring transferability and in quantifying usability in a multiple device context. This study also answers the research questions regarding the impact of task complexity, user experience, and device order on the total usability of the system. Study one follows a systematic approach to develop, validate, and apply a new questionnaire tailored specifically to measure the transferability within a multiple device system. The System Transferability Questionnaire (STQ) is obtained after validation with 15 question items. In a software usability study, the STQ demonstrated excellent internal reliability and validity. Results show that the STQ is effective in capturing four factors regarding transferability, which are transfer experience (TE), overall experience (OE), consistency perception (CP) and functionality perception (FP). Validation results show good convergent, discriminant, criterion and nomonlogical validity. Study two adopts a systematic tool to consolidate usability constructs into a total usability score. The study utilizes principal component analysis (PCA) to determine the weight of the four usability components (satisfaction, transferability, effectiveness, and efficiency), which is used when obtaining the total usability score. Results show slightly different weights for the four components. This quantitative tool can be applied in different usability context in which multiple devices are involved. Usability specialists are encouraged to adjust the tool based on different usability scenarios. Study three investigates the impact of task complexity, user experience, and device order on the total system usability. Results show that the total usability score is not affected by task complexity, user experience or device order. However, lower physical task complexity leads to longer performance time and lower errors from the users. High experienced users have significantly lower errors made in tasks. The machine order also has divergent results. When the mini-lathe machine was used first, users had better transferability results but poorer performance outcomes as compared to when the drill press was used first.

The purpose of this study was to determine if task type, task complexity, and search mechanism would have a significant affect on task performance. The problem motivating this study is the potential for systems online help designers to construct systems that can improve the performance of computer users when they need help.

This thesis highlights collaboration among secondary school teachers as an example of employees' possibilities to learn from each other in collaboration. The purpose of the thesis is to throw light upon how employees shape conditions for cooperation, and the importance these conditions have for their possibilities to exchange experiences and influence each other’s task understanding, i.e., their possibilities to learn from each other through collaboration. The thesis’ theoretical bases are action theory and constructivism. The study uses a case-study approach, and involves a combination of participant observation and interviewing techniques. The starting point was to identify factors that appeared to assist or hinder the exchange of experience. One conclusion is that the division of labour seems to have been central to how conditions are shaped and is the basis for teachers' possibilities for experience-exchange. The combination of a common, physical context, and the friction between the teachers’ actions, seems to promote experience-exchange between teachers and contribute to the possibility to learn from each other. Friction could be seen as the basis for teachers’ need to achieve consensus and coordinate action, since friction is a hindrance. Consequently, friction entails both obstacles and opportunities for coordination and learning. The conditions giving rise to frictions that are overcome seem to be shaped by the way teachers divide the work. Tasks with a high degree of complexity seldom cause friction, which is avoided in these contexts through individual labour. Friction seems to occur and be overcome more frequently during cooperation on easier tasks. Teachers' attempts to collaborate on complex tasks, and simultaneously divide work individually, seem to increase the risk of insurmountable friction arising, from power-related and communication barriers.

The topics of motivation and job satisfaction have been of interest to researchers over the past decades. Many researchers and practitioners have studied the driving factors to motivate staff in the workplace, but no specific findings have shown correlation between motivation and job satisfaction. Problems of motivation and job satisfaction have continued to plague many developing countries like Malaysia. This study is an attempt to improve the understanding of the human contribution to variability in manufacturing industries and the focus areas are work motivation, satisfaction and performance as well as relationships with learning behaviours of employees in the workplace. The research work can be categorized into three parts. The first part consists of pilot study which was conducted to determine the practicality and validity of method/instruments used in the research. The pilot study also helped to correct the flaws/weaknesses of the method before employing it in the full-scale research study. Secondly, an experimental study was carried out to identify the motivation, satisfaction, performance and learning behaviour of unskilled and skilled employees doing simple or complex tasks individually or in group. The final part was an industrial study conducted with 356 employees from various positions and backgrounds in selected manufacturing industries in Malaysia. Based on the findings, it has been shown that unskilled employees preferred doing complex tasks in a group rather than doing simple tasks and skilled employees preferred doing complex tasks individually rather than in a group. It increased their work motivation, satisfaction and performance. It was found that task identity (simple tasks) and learning behaviours (individual learning of unskilled employees) can be the reason for employees to leave in the future. The task identity (task complexity) can be an important factor in job design in organisations and it is significant in the learning process of unskilled and skilled employees in manufacturing industries, particularly in Malaysia. It was also found that learning in a team (group) appears to be a very significant factor in workplace learning for both unskilled and skilled employees. The study has shown that there are relationships between motivational and learning behaviours of skilled and unskilled employees and this knowledge is expected to be useful for employers and policy makers in organisations especially in manufacturing industries in Malaysia.

The need to develop appropriate computer training techniques has led to an explosion of research in information systems. One of the most recent studies was conducted by Compeau and Higgins (1995) in which two training methods were examined in the context of Social Cognitive Learning Theory (Bandura, 1977, 1978). The training methods examined were traditional lecture-based training and behavior modeling. Based on various anomalies found in the Compeau and Higgins study, this paper introduced a moderating variable, task complexity, into their model and also attempted to replicate their original experiment. This study also incorporated an additional training method, Computer Aided Instruction (CAI), which was examined in an experiment by Gist, Schwoerer, and Rosen (1989). It was hypothesized that task complexity has a moderating effect on the relationships between behavior modeling and performance, between behavior modeling and self-efficacy, and between self-efficacy and performance. Finally, an empirical investigation was performed to determine the relative effectiveness of the three training methods examined. To test these hypothesized relationships, an experiment was conducted that examined prior performance, self-efficacy, outcome expectations, and actual performance at two levels of task complexity for each of the three training methods. The data were analyzed using a combination of multivariate and univariate analyses of variance and a structural equation modeling software package, AMOS©. Five of the original nine hypotheses from the Compeau and Higgins study were fully supported; however, none of the task complexity and only one of the avoidance behavior hypotheses were supported. Possible causes of this lack of support were multi-dimensionality of constructs or the need to examine task dimensions other than complexity. Relevant findings in this study included (1) a positive significant relationship between behavior modeling and final performance, (2) a positive significant relationship between prior performance and the endogenous constructs in the model: computer self-efficacy, outcome expectations, and final performance, and (3) a ranking of the three training methods in terms of effectiveness. Although behavior modeling produced the best performance results at all levels of task complexity, CAI was equally effective when the level of complexity was high. For low complexity tasks, however, CAI was the least effective method examined. / Ph. D.

Guided by the research on cognitive performance tasks (e.g., Wood, Mento, & Locke, 1987), the potential moderating effects of skill and outcome difficulty upon the relationship between specific, difficult goals and physical task performance were examined from an attribute treatment approach. Overall, different measures of performance yielded several convergent findings. Using a sample of 184 laboratory participants, a test of the primary hypothesis failed to support the expected three-way interaction among skill, outcome difficulty and goal setting. Models containing skill and outcome difficulty were found to provide the most parsimonious explanation of variance in performance, regardless of whether assigned or personal goal were used as an index of motivation. After controlling for skill and outcome difficulty, goal commitment showed a weak positive relationship with task performance. Self-efficacy was not related to performance among participants pursuing specific, difficult goals. Potential reasons for the failure to obtain evidence of the predicted three-way interaction among skill, outcome difficulty, and goals are discussed in the context of limitations to the present study. Future research directions are suggested. / Master of Science

Introdução: Ainda não é conhecido na literatura se os sujeitos pós-AVC têm a capacidade de aprender ações complexas. Os estudos existentes apresentam falhas metodológicas, assim como na implementação da complexidade da tarefa. Este trabalho investigou os efeitos da manipulação da complexidade da tarefa na aprendizagem motora desta população. Método: Vinte e quatro sujeitos pós-AVC (Grupo experimental - GE) e vinte e quatro sujeitos saudáveis (Grupo controle - GC) foram selecionados e divididos em dois experimentos: baixa complexidade e alta complexidade. Foram desenvolvidas duas tarefas denominadas de baixa e de alta complexidade, com variação não só no número de elementos, mas também na carga de processamento exigida para a execução da mesma, as quais deram origem ao experimento 1 e 2, respectivamente. Esta tarefa era executada em ambiente de realidade virtual, a partir do deslocamento do centro de pressão para controle dos objetos móveis na tela. O delineamento foi constituído de 150 tentativas para a fase de aquisição, dividida em 3 dias de prática. Após 4 dias sem prática foi realizado o teste de retenção (RET) e transferência (TR). A alteração realizada no teste de TR foi na direção do deslocamento dos objetos. As variáveis dependentes de ambos experimentos foram: pontuação e tempo de execução. Para cada variável dependente foi conduzido uma Anova two-way com medidas repetidas (2 grupos x 4 momentos) seguido de post hoc de Tukey para identificar os momentos das mudanças. Resultados: No experimento 1, para a variável pontuação ambos os grupos melhoraram o desempenho ao longo da aquisição e o mantiveram no teste de retenção. Não houve diferença entre o GC e GE. Para a variável tempo de execução, o GC diferenciou-se do GE em todos os momentos (Início e final da aquisição, teste de retenção e transferência), apresentando pior desempenho. No experimento 2, tanto para a variável pontuação, quanto para tempo de execução o GE diferenciou-se do GC no final da aquisição e no teste de RET. Conclusão: Os sujeitos pós-AVC foram capazes de aprender a tarefa de baixa complexidade, mas esta aprendizagem não foi passível de generalização. Quanto a tarefa de alta complexidade, os sujeitos do GE não aprenderam a tarefa, sendo muito impactados pelo aumento de complexidade. Possivelmente, os déficits cognitivos relacionados ao planejamento e sequenciamento de ações impactaram na aquisição da habilidade de alta complexidade e não prejudicaram a aprendizagem da tarefa de baixa complexidade / Introduction: It is not yet known in the literature wether post stroke subjects have the ability to learn complex actions. The studies have methodological flaws as well as the implementation of the complexity of the task. This study aimed to investigate the effects of manipulation of task complexity in motor learning in this subjects. Method: Twenty-four post stroke subjects (experimental group - EG) and twenty-four healthy subjects (control group - CG) were selected and divided into two experiments: low complexity and high complexity. Two tasks were developed, low and high complexity, with changing not only on the number of elements, but also on the processing load required, which led to the experiment 1 and 2, respectively. This task was performed in a virtual reality environment, considering the displacement of the center of pressure for controlling the objects move on the screen. The design consisted of 150 trials for the acquisition phase, divided into three days of practice. After 4 days was performed retention test (RET) and transfer (TR). The TR test included the change in the direction of the objects movement. The dependent variables of both experiments were: score and runtime. For each dependent variable was conducting a two-way ANOVA with repeated measures (2 groups x 4 times) followed by Tukey post hoc. Results: In experiment 1, in regards to the variable score, both groups improved performance and kept in the retention test. There was no difference between the CG and EG. For the variable runtime, the CG differed from EG at all times (Beginning and end of the acquisition, RET and TR), with worse performance. In Experiment 2, both the variable score runtime, the EG differed from the CG at the end of acquisition and RET test. Conclusion: The post-stroke subjects were able to learn a low complexity task, but this learning was not generalized. As the higher complex task, post stroke subjects have not learned the task, being very impacted by the increase of complexity. Possibly the cognitive deficits related to planning and sequencing actions impacted the acquisition of the higher complex skill and not affected the low complexity task learning

O objetivo desse trabalho foi investigar os efeitos da estrutura de prática no processo adaptativo de aprendizagem motora em razão da complexidade da tarefa. Os participantes foram 160 crianças de ambos os gêneros e a tarefa foi de timing coincidente. No experimento 1, as crinaças executaram uma tarefa que consistiu tocar certos alvos de forma seqüencial em integração a um estímulo visual (tarefa simples) e, o experimento 2, uma tarefa com 6 toques (tarefa complexa). O delineamento dos dois experimentos constou de quatro grupos (prática constante, prática aleatória, prática constante-aleatória, e prática aleatória-constante) e de duas fases de aprendizagem (estabilização e adaptação) As variáveis dependentes foram os erros absoluto, variável e constante. Os resultados mostraram que a complexidade da tarefa afetou os efeitos da estruturação da prática no processo adaptativo de aprendizagem motora. Verificou-se que, para a aprendizagem da tarefa simples, a adaptação ocorreu com similar desempenho para os quatro grupos, enquanto que para a tarefa complexa, o desempenho foi pior para o grupo constante / The objective of this work was to investigate the effects of practice schedule in the adaptive process of motor learning in reason of task complexity. The participants were one hundred and sixty children of both gender and the task was of coincident timing. In the experiment 1 the children performed a task that consisted of touching four response keys sequentially in conjunction with a visual stimulus (simple task), and, the experiment 2, the task evolved six touches (complex task). The experimental design of both experiments consisted of four practice groups (constant, random, constant-random and random-constant) and two learning phases (stabilization and adaptation). Dependent variables were absolute, variable and constant errors. The results show that task complexity affected the effects of practice schedule in the adaptive process of motor learning. It was verified that for learning of simple task the adaptation occurred with similar performance for all groups, while the complex task occurred worse performance for constant group

Introdução: Ainda não é conhecido na literatura se os sujeitos pós-AVC têm a capacidade de aprender ações complexas. Os estudos existentes apresentam falhas metodológicas, assim como na implementação da complexidade da tarefa. Este trabalho investigou os efeitos da manipulação da complexidade da tarefa na aprendizagem motora desta população. Método: Vinte e quatro sujeitos pós-AVC (Grupo experimental - GE) e vinte e quatro sujeitos saudáveis (Grupo controle - GC) foram selecionados e divididos em dois experimentos: baixa complexidade e alta complexidade. Foram desenvolvidas duas tarefas denominadas de baixa e de alta complexidade, com variação não só no número de elementos, mas também na carga de processamento exigida para a execução da mesma, as quais deram origem ao experimento 1 e 2, respectivamente. Esta tarefa era executada em ambiente de realidade virtual, a partir do deslocamento do centro de pressão para controle dos objetos móveis na tela. O delineamento foi constituído de 150 tentativas para a fase de aquisição, dividida em 3 dias de prática. Após 4 dias sem prática foi realizado o teste de retenção (RET) e transferência (TR). A alteração realizada no teste de TR foi na direção do deslocamento dos objetos. As variáveis dependentes de ambos experimentos foram: pontuação e tempo de execução. Para cada variável dependente foi conduzido uma Anova two-way com medidas repetidas (2 grupos x 4 momentos) seguido de post hoc de Tukey para identificar os momentos das mudanças. Resultados: No experimento 1, para a variável pontuação ambos os grupos melhoraram o desempenho ao longo da aquisição e o mantiveram no teste de retenção. Não houve diferença entre o GC e GE. Para a variável tempo de execução, o GC diferenciou-se do GE em todos os momentos (Início e final da aquisição, teste de retenção e transferência), apresentando pior desempenho. No experimento 2, tanto para a variável pontuação, quanto para tempo de execução o GE diferenciou-se do GC no final da aquisição e no teste de RET. Conclusão: Os sujeitos pós-AVC foram capazes de aprender a tarefa de baixa complexidade, mas esta aprendizagem não foi passível de generalização. Quanto a tarefa de alta complexidade, os sujeitos do GE não aprenderam a tarefa, sendo muito impactados pelo aumento de complexidade. Possivelmente, os déficits cognitivos relacionados ao planejamento e sequenciamento de ações impactaram na aquisição da habilidade de alta complexidade e não prejudicaram a aprendizagem da tarefa de baixa complexidade / Introduction: It is not yet known in the literature wether post stroke subjects have the ability to learn complex actions. The studies have methodological flaws as well as the implementation of the complexity of the task. This study aimed to investigate the effects of manipulation of task complexity in motor learning in this subjects. Method: Twenty-four post stroke subjects (experimental group - EG) and twenty-four healthy subjects (control group - CG) were selected and divided into two experiments: low complexity and high complexity. Two tasks were developed, low and high complexity, with changing not only on the number of elements, but also on the processing load required, which led to the experiment 1 and 2, respectively. This task was performed in a virtual reality environment, considering the displacement of the center of pressure for controlling the objects move on the screen. The design consisted of 150 trials for the acquisition phase, divided into three days of practice. After 4 days was performed retention test (RET) and transfer (TR). The TR test included the change in the direction of the objects movement. The dependent variables of both experiments were: score and runtime. For each dependent variable was conducting a two-way ANOVA with repeated measures (2 groups x 4 times) followed by Tukey post hoc. Results: In experiment 1, in regards to the variable score, both groups improved performance and kept in the retention test. There was no difference between the CG and EG. For the variable runtime, the CG differed from EG at all times (Beginning and end of the acquisition, RET and TR), with worse performance. In Experiment 2, both the variable score runtime, the EG differed from the CG at the end of acquisition and RET test. Conclusion: The post-stroke subjects were able to learn a low complexity task, but this learning was not generalized. As the higher complex task, post stroke subjects have not learned the task, being very impacted by the increase of complexity. Possibly the cognitive deficits related to planning and sequencing actions impacted the acquisition of the higher complex skill and not affected the low complexity task learning