This is the third section of my post on BBST 4.0. The other parts are at:
- 1. Background: What is BBST. (If you are already familiar with BBST, skip this)
- 2. Differences Between the Core BBST Courses and Domain Testing
- 4. What We Think Should Change
- 5. Financial Model & Concluding Thoughts
Learning Objectives and Structure of Foundations 3.0 (2010)
Most courses present content, help students develop skills, and foster student attitudes. The typical course identifies its central topic area (such as software testing or Java programming) and focuses on that, but also presents other content, skills and attitudes that are not directly on-topic but support the student who studies this area. For example, we see communication skills as important for software testers.
In Foundations 2.0 and 3.0, we decided that the central testing content should be the five most important issues in software testing. In our view, these were:
- Information objectives drive the testing mission and strategy (Why do we test? What are we trying to learn and how are we trying to learn it?)
- Oracles are heuristic (How can we know whether a program passed or failed a test?)
- Coverage is a multidimensional concept (How can we determine how much testing has been done?)
- Complete testing is impossible (Are we done yet?)
- Measurement is important, but hard (How much testing have we completed and how well have we done it?)
Along the way, to teach these, we have to present definitions and foundational information.
Along with the central topic, we worked on a key set of supporting objectives:
- Help students develop learning skills that are effective in the online-course context. Most professional-development courses are easy. Even if they present difficult material, they give the students feel-good activities that help the student feel as though they understand the material or can do the basic task, whether they actually learned it or not. This is good marketing (the students walk away feeling successful and happy) but we don’t think it’s good education. We prefer to give students tasks that are more challenging, that helps them understand how well they have learned the content and how well they can apply the skills. Most of our professional students finished their formal schooling years ago and so they are out of practice with academic skills. We have to help them rebuild their skills with effective reading, taking exams, writing assignments, providing peer reviews and coping with critical feedback from peers and instructors.
- Foster the attitude that our assessments are helpfully hard without being risky
To be instructionally effective, assessments (exams, quizzes, labs, projects, writing assignments, etc.) should be hard enough that students have to stretch a bit in order to do them well, but not too hard or students will give up.Students come into Foundations with vastly different backgrounds. This is not like a third-year math course, in which the instructor knows that every student has two successful years of math behind them. Our hardest assessment-design goal is to create activities that are challenging for experienced professionals while still being motivating and instructionally useful for newcomers to the field.One of the course-design decisions that comes out of an intention to serve a very diverse group of students is that the assessments should be challenging, feedback should be honest but pass-fail decisions should be generous. Honest feedback sometimes requires direct and detailed criticism. Our goal is to deliver the criticism in a kind way. We are trying to teach these students, not demoralize them. While this is not possible with every student, the goal is to write in a way that helps them understand what is meant, rather than making them so defensive or embarrassed that they reject it.
- Our intention is that most students who take Foundations and try reasonably hard should “pass” the course. Most of the people who don’t pass the course should be people who abandon it, often because they had to switch their focus to some crisis at work. Someone who has to shift focus (and stop working on BBST) won’t pass the course, but we don’t want to say they “failed” it. That’s not what happened. They didn’t fail. They just didn’t finish. So rather than saying that someone “passed” a BBST course, we say they completed it. Rather than say they “failed” our course, we say that they didn’t complete it.
- Some students stay in the course to the end but don’t do the work, or do work that is really bad, or submit work that isn’t theirs (otherwise known as “cheating.”) These students don’t complete the course.
- Our standard, for students who make an honest effort, is this: Bug Advocacy and Test Design are a little harder than Foundations. If, at the end of the course, the instructor believes that a student has no hope of succeeding in the next class, we won’t tell them they “completed” Foundations.
- Most students who stay in the course do get credit for completing the course even if their work is not up to the average level of the class. They get honest feedback on each piece of work, but if they make an honest effort, we want to give them a “pass.”
Bug Advocacy and Test Design emphasize other supporting objectives, such as persuasive technical communication, time management and prioritization, and organization of complex, multidimensional datasets with simplifying themes. However, for students to have any hope with these, they need the basic learning skills that we try to help them with in Foundations.