Computational Thinking - A Practical Guide to Teaching Computing and ICT in the Secondary School - page 97

should be applied. It is context-dependent and, in investigating different possible
solutions, will vary. Any element can also carry greater or equal weighting.
There are a number of techniques pupils need to be able to use and understand
in order to engage in Computational Thinking. These include: creating algorithms
and flowcharts; testing and debugging.
Decomposition
When looking at a problem it can help to break it down into a series of parts or
subproblems. For example, the process of making dinner could be subdivided into
the following activities: make starter; make main course; make sweet. This process,
in Computational Thinking, is called 'decomposition'. Pupils need to understand
that it allows the problem to be solved in separate or incremental stages, or to be
solved by different people or groups at different times. It helps to make complex
processes more manageable.
The organisation of data can also be decomposed. For example, in Geography,
information about the population of a country can be decomposed into entities,
such as gender, occupations, places of residence, etc. This will allow for more
meaningful outcomes to emerge from any analysis.
Sometimes the subprograms themselves can also be decomposed further.
However, it is important that pupils understand not to continue with deconstruction
to a point where coherence is lost. Another important point to teach pupils is that
by decomposing into subproblems there is a potential danger if any one of the
solutions to the subproblem is found to be incorrect, then the whole, when
recomposed, will be incorrect. A classic example to illustrate this is that Charles
Babbage, a nineteenth-century mathematician, amongst other things, was aware
that tables of information (tide, movement of the planets) were created by teams of
people called 'computers' (one who computes). Their task was to undertake a small
part of the overall complex calculation associated with the area on which
information was being compiled. These parts, identified by decomposing the
original problem into much smaller calculations, would be answered, and the
solutions gathered together to provide an overall solution. Small errors or
inaccuracies in the subdivisions, of which there were many, were amplified once
they were recomposed. This led to significant errors in the final tables, and much
frustration by sailors and astronomers. Babbage attempted to devise a machine to
do these calculations, instead of relying on people. Some regard this as the origin
of the modern computer.
Pattern Recognition
If the problem, or the subproblem, being considered is based on or is similar to a
previous problem, then the element of Computational Thinking known as 'Pattern
Recognition', can be helpful. This is particularly true if there is significant data
associated with the prior problem that has been gathered over time. In Pattern
Recognition, the data is looked at to see if there are patterns or trends. Plotting
graphs, establishing lines of best fit, creating charts et al. are useful to help with
this. If patterns are identified by pupils, these can help with hypothesising what
might now happen. For example, people looking for trends in the performance of
shares can use them to help them to decide when to buy and sell.