EP = Equivalence Partitioning. As an example, if you have a range of valid values, like 1-10, you would choose to test one valid value (say 7), and one invalid value (like 0).
BVA = Boundary Value Analysis. If you take the example above, you would test the minimum and maximum boundaries (1 and 10), and test beyond both boundaries (0 and 11). Boundary Value Analysis can be applied to a field, record, file, or anything with a stated or implied limit of some kind.
CE= Cause/effect. This is normally input of a combination of conditions (cause) in order to yield a single system result or transformation (effect). For example, you might want to test the ability to add a customer using a particular screen. This may involve entering multiple fields, such as name, address, and phone number, followed by pressing the “add” button. This is the “cause” portion of the equation. Once you press the “add” button, the system will return a customer number and add the customer to the database. This is the “effect”.
EG = Error guessing. This is when the test analyst uses their knowledge of the system and ability to interpret specifications to “guess” at what type of input might yield an error. For example, perhaps the spec says “the user must enter a code”. The test analyst will think “what if I don’t enter a code?”, “what if I enter the wrong code?”, and so on. This is error guessing.
ECP = Equivalence Class Partitioning – A software testing technique that involves identifying a small set of representative input values that invoke as many different input conditions as possible.