Multivariate Testing

The attention span on the Web has been decreasing ever since Google had arrived and changed the rules of the game. Now with millions of results available on any topic imaginable, the window to grab a visitor’s attention has decreased significantly (in 2002, the BBC reported it is about 9 seconds)


Statistical testing relies on design of experiments. Several methods in use for multivariate testing include:

1. Discrete choice and what has mutated to become choice modeling is the complex technique that won Daniel McFadden the Nobel Prize in Economics in 2000. Choice modeling models how people make trade offs in the context of a purchase decision. By systematically varying the attributes or content elements, one can quantify their impact on outcome, such as a purchase decision. What is most important are the interaction effects uncovered, which neither the Taguchi methods nor Optimal design solve for.
2. Optimal design involves iterations and waves of testing. Optimal design allows marketers the ability not only to test the maximum number of creative permutations in the shortest period of time but also to take into account relationships, interactions, and constraints across content elements on a website. This allows one to find the optimal solution unencumbered by limitations.
3. Taguchi methods: with multiple variations of content in multiple locations on a website, a large number of combinations need to be statistically tested and medium/low traffic websites can take some time to get a large enough sample of visitors to decide which content gives the best performance. For example, if 3 different images are to be tested in 3 locations, there are 27 combinations to test. Taguchi  orthogonal arrays can be used in the design of experiments in order to reduce the variations but still give statistically valid results on individual content elements. Taguchi uses fractional factorial designs.

The principal goal of education

The principal goal of education is to create men who are capable of doing new things, not simply of repeating what other generations have done. Jean Piaget (1896-1 980)


Piaget (1972) argues that it is better to let children spend more time on a few problems, really working through them, than to cover a lot of territory: "It  is in learning to master the truth by oneself at the risk of losing a lot of time and of going through all the roundabout ways that are inherent in real activity" (p. 104). Early critics of Piaget faulted him for basing his worldview of children on studies done in his own home. Subsequent research by  others, however, seemed to indicate that his creative genius made it possible for him to do successfully what others would frown on. Today, early education is strongly influenced by Piaget, particularly when
we put down the skillpacks and dittos in favor of less directive hands-on learning

E-Learning is an approach to teaching -1

eLearning, I suppose, is an  approach to teaching as opposed to being a  tool of teaching. Of course, eLearning involves the use of a number of technological tools. However, eLearning, as an approach, primarily, to encourage students to construct their own knowledge in collaboration with other learners as well as experts in that subject matter. eLearning, therefore, is about collaborative construction of knowledge.

The use of technology in education, on the other hand, has a significant history. Initially, computers were applied in behaviourist modes in accordance with Skinner’s work (Ravenscroft 2001), which emphasized the teacher’s control over what is learned and how it is to be learned.

However, the emphasis, more recently, is on the co-construction of knowledge in collaboration with peers and/or  others which provide students with opportunities to develop their own understandings. Skinner’s behaviourism, Piaget’s cognitive constructivism and Vygotsky’s social constructivism can all be facilitated through eLearning. Tools and technologies are only a part of the whole issue. eLearning primarily is an approach to teaching.

Process & Procedures

'A process is a systematic series of actions directed towards the achievement of a goal'(Joseph Juran). Processes are one of the 'Three P's' of quality in business, People, Products and Processes, where products are what is done, and processes are about how it is done. Thus people use processes to produce products. The word 'products' here includes the end goal of all actions, which is often an intangible service rather than a more tangible product. Thus statements about quality often talk about 'products or services' rather than plain 'products'. A way of contrasting quality in product and process activities is by viewing production of quality products as doing the right thing and quality processes are about doing the thing right, as in the diagram below.

Reference:

Consumerism and Product Quality", Quality Progress (New York, New York: American Society for Quality Control), July, 1970

Thinking Software Quality

Some Definitions:

1. "Software Quality is the fitness for use of the software product" [1]
2. Kitchenham states quality is "Hard to define, impossible to measure, easy to recognize" [2].
3. Gilles states, "Quality is generally transparent when present, but easily recognized in its absence" [3].

While quality in theory can be defined, in practice and use, an absolute definition is elusive. This is part of the knowledge that needs to be captured - how to apply these abstract SQA concepts to real projects. Here are some lessons:

1. Project Managers and Software Developers need to understand what "Software Quality Assurance" is and how their project can benefit by its application. 
2. Software Quality Assurance implementation is a balancing activity that must be tailored as project appropriate. 
3. Software Quality Assurance must evaluate the process as well as the products. 
4. There must be a Software Assurance Plan.
5. Software Quality Assurance must span the entire Software Development Life Cycle.
6. Requirements, the birthplace of successful projects.
7. Software Quality Assurance does NOT Equal Testing.
8. Metrics are a necessity.
9. Safety and Reliability are critical aspects of SQA.
10. Independent Verification and Validation (IV&V) is an important tool within SQA.
11. Hardware Does NOT Equal Software!
12. Risk Management is NOT Optional.

References
[1] Schulmeyer, G. Gordon and McManus, James I., Handbook of Software Quality Assurance, 3rd Edition, Prentice Hall PRT, 1998
[2] Kitchenham, Barbara, Pfleeger, Shari Lawrence, "Software Quality: The Elusive Target", IEEE Software, Vol 13, No 1 (January 1996) 12-21
[3] Gillies, Alan C., Software Quality, Theory and Management, International Thomson Computer Press, 1997