An accessor is a type of method used in object-oriented programming languages to return the value of an data member. (Compare with mutator). Public accessor methods are a best practice, recommended by many coding standards as the correct way to access fields of a class, while the member itself can remain private.

Some conventions even mandate that access within the class itself should be through the accessor methods. This ensures that attempts to read the instance variable are all performed at a single point in the code, which can often be of assistance when debugging multi-threaded applications.

The format of an accessor method is uncomplicated almost by definition. For example, given a member variable "balance" of type double, the corresponding accessor would be:

public double getBalance()
    return balance;

While most accessors take no arguments, on occasion there is value in an accessor that takles an index to return a single element of an array or collection, rather than the entire collection itself. Design patterns such as JavaBeans make use of this kind of convention, and treat such accessors as an indexed property.

Accessors are by far the most important type of setter method. However in many object-oriented languages such as Java, they do present some issues when they are used to access member variables that are themselves objects. The member is returned as a reference, which may allow the caller to inadvertently modify it. This is particularly an issue with collection classes. In other languages such as C++ that support the notion of const, it is a good idea to make an accessor a const method that returns a const reference.