Page 469 - confbook2022

The Tenth International Arab Conference on Quality Assurance in Higher Education ﻲﻟﺎﻌﻟا ﻢﯿﻠﻌﺘﻟا ةدﻮﺟ نﺎﻤﻀﻟ ﺮﺷﺎﻌﻟا ﻲﻟوﺪﻟا ﻲﺑﺮﻌﻟا ﺮﻤﺗﺆﻤﻟا

Measures needed. Faults/KLOC, the number of faults divided by the number of lines of code. Completeness, the
number of features completed divided by the number of features required. Students Understanding: a pre and post
quiz analyzed by T-Test or Mann-Whitney.

Experiment design. With a view to generalize the results, subjects will be chosen using simple random sampling as
a probability sampling technique, where they are selected randomly and optionally from the SWE course students at
the university. The size of the sample is: 8 groups with a total of 38 student. In addition to randomization, if any
group out of the 3rd year engaged the experiment we do blocking on prior experience by a pre-test, and on differences
between problems they worked on, in terms of complexity and context. Moreover the experiment uses a balanced
design, which means that we have the same number of subjects per treatment in the design. The instruments for
performing and monitoring the experiment are: objects represented by real products and its requirements, guidelines
and principles for the development methodologies. And to guarantee rightful comparison, we concerned comparable
support of available resources and training for the two methods for the teams. Lastly, measurements instrumentation
conducted via data collection in manual and online forms and interviews and some reads from some tools used.

1.8 Operation:

We select and inform participants about the experiment nature, and prepare material i.e. forms and tools, then ensure
the infrastructure needed.
The course is designed to fulfil the Pyramid of Agile Competences described in [6]. Each group have name and
logo to pace up work by adding a kind of competition. The project is implemented during seven 2-weeks sprints.
Lectures differ between discussion lectures, gaming, and technical workshops.
Several tools have been integrated, mainly which are easy, less-cost, compatible and interrelated tools to use from
the same screen. We use “Trello” for Project Management, it is a platform supports the transparency idea in Scrum.
Through it teams organize the backlog, and tasks of the current sprint and show their progress and communicate. It
is compatible and powered up by other essential applications. We also use “Bitbucket” for source code management,
“Catme” for team work assessment, “Webinar” for online meetings with lecturer, and “Moodle” and online forms
for quizzes and questionnaires.
Roles assigned with their self-explanatory names [1]. Scrum Master initially assigned to the teacher, then a student
chosen according the sprint details [5], or voted [6]. Scrum Coach was external senior developer. Product owner
assigned to students. Others became developers and testers. Finally teacher’s role was expressing the “guide on the
side, not a sage on the stage” principle [14].
A formative and summative evaluation assess students understanding, skills, projects and satisfaction. Also
monitoring by observations and evaluations, written down into dedicated tables.

Results and Discussion:

Projects were of the same context and level of complexity, but different products and details. Below in Table 1 is
a summary of teams and projects.

Table 1. Summary of teams and projects.
type\method Agile Waterfall Project Context
Desktop First team Fifth Team Dental Clinic
"Hash the Dash" Systems
Second team Sixth Team Medical Laboratory
“Alpha Team” Systems
Third Team Seventh Team Cars Insurance
“Rainbow Team” Systems
Web Fourth Team Eighth Team Child Care Center
"CCCare Team" Systems
1.9 Evaluation of Learning Outcomes:

For the Pedagogical model factor, the difference between the pre and post quizzes results were taken to find the
improvements, which were about 44% for the Agile students against roughly 28% for the traditional on SPSS
software.

٤٤٨