Computational Thinking (CT)

Computational Thinking (CT) is a thinking process used to solve problems logically and methodically.
Computational Thinking promotes a purposeful, describable, and replicable method of problem solving.

Steps in Computational Thinking

Decomposition

Breaking a problem into smaller pieces, or subprocesses, to

describe, understand, or better solve the problem.
Pattern Recognition
Identifying similarities or common differences that help us
describe patterns and make predictions.

Pattern Generalization and Abstraction

Organizing relevant information and recognizing commonalities

between patterns to make one solution work for multiple
problems.
Algorithm
A process, or list of steps, followed to solve a
problem or complete a task.

Decomposition

The computational thinking process begins with the decomposition of the problem—breaking down the
process into a set of smaller subprocesses that allows students to describe, understand, or better execute
the process. Decomposition includes dividing a task into a sequence of subtasks or identifying elements or
parts of a complex system. For example:
• Providing directions to a location.
• In mathematics, numbers are decomposed in a variety of ways.
o 247 is 200 + 40 + 7
o 247 is 2(100) + 4(10) + 7
o 247.82 is 2 * 102 + 4 * 101 + 7 * 100 + 8 * 10 -1 + 2 * 10 -2

Pattern Recognition

Once a problem is decomposed, similarities or common differences are identified that will help in making
predictions or discovering shortcuts. This step is called pattern recognition. Examples include:
• Shoppers looking for the shortest line at the supermarket when choosing a register for
check out or when watching the stock market to decide when to buy and sell stocks. Drivers
also look for patterns when driving in traffic, helping them decide if and when to change
lanes. Scientists look for patterns in data to derive theories and models.
 Pattern Generalization and Abstraction

After identifying patterns, the next step is to abstract and generalize the patterns. In this process, the
relevant information for the problem being solved is kept and irrelevant information is dismissed.
• Abstraction is used to organize or categorize things (a human is a mammal, a mammal is an
animal, and so on). Abstraction reveals the "big picture", so reasoning can occur without
thinking about details. Abstraction allows learners to transfer learning or learn by creating
an analogy. For example:
o A summary is given when telling a story or describing a movie to a friend; not every
detail is provided.
o A world map is an abstraction of Earth in terms of longitude and latitude, helping to
describe the location and geography of a specific place.
o A book report is an abstraction because it is a summary only covering the theme or
key aspects of the book.
• When learners generalize patterns, they identify common or shared characteristics between
two domains or problems so that models or solutions of one can be adapted or applied to
the other. When studying trends, norms, outliers, and scalability, patterns are used to
generalize. For example:
o When studying mammals, learners know they are warm-blooded, give live birth, have
hair, etc. An elephant is a mammal. Therefore, the elephant is warm-blooded, gives
live birth, has hair, etc.
o Pattern generalization is utilized when using a search engine. Pattern generalization
identifies and suggests popular keywords, such as through auto-complete or
autocorrect. Online shopping companies use pattern generalization to predict what
will be of interest to their customers and make recommendations accordingly.

Algorithm Design

The final step is to design an algorithm or develop a step-by-step strategy for solving a problem.
Algorithmic thinking involves both the creation and the execution of an algorithm. For example:
• When a grandmother writes down a family recipe for her signature dish, an algorithm is
created that others can follow to replicate the dish.
• When using online maps to get directions to a friend’s house, there is a specific sequence of
steps to follow.
• A science lab experiment has a set of instructions to carry out to collect and analyze data.

Once an algorithm is designed, it is essential to evaluate the solution for correctness. The algorithm
should meet all the requirements to solve the problem and the algorithm should perform with efficiency
and reliability. Optimizing algorithms is critical because solutions should be usable, efficient, reliable, and
effective. If a solution can be solved with 9 steps instead of 14, the algorithm should be optimized for
efficiency. When folding a paper airplane, a test is completed to check its ability to fly. Then modifications
or revisions might be made to the design to improve the outcome. All of these demonstrate optimization
and evaluation of a solution.