Identification keys are tools for helping identify an unfamiliar object. Biologists use them to identify organisms. Identification keys require users to make a series of decisions about the organisms they wish to identify. Each decision will eliminate some possibilities determine what decision will need to be made next if the first decision did not eliminate all but one possibility. This site will discuss text-based keys of three kinds: directed, undirected, and semi-directed. Few parts of the world have enough images to develop effective image-based keys. Moreover, text-based keys help students master their discipline's vocabulary.
In directed keys, the author determines the decisions users need to make. They have no choice. Most directed keys are dichotomous. This means that users always have to decide between two alternatives. There are some directed keys in which the number of alternatives at each step varies, but they are exceptional. Directed keys for a small number of taxa are easy to write. Asking student to write keys for specimens on their desk will help them learn in many ways, including helping them understand how to use a dichotomous key. A sample lesson for teaching students to write a key is available as a page or pdf.
Undirected, or “multi-access” keys require the user to decide which of many features to use at each step. The advantage is that users can select the features that are present on their specimen and that they understand. As with all keys, each decision will eliminate some possibilities. Each successive decision eliminates more of the remaining possibilities until finally only one is left. Undirected/multiaccess keys can make identification much easier, particularly if the sample being identified does not have some of the features required by a dichotomous/directed key, they are difficult and time-consuming to prepare. An example of a multi-access key is here.
Semidirected keys present users with several choices at each step, but not all the possible choices. The first choice will help the user identify the major group of plants to which the sample belongs. This choice will lead to another set of choices. The program will present those that, based on the previous choice and the data available in the database, will narrow down the answer most rapidly. There are few semi-directed keys available, but Symbiota uses this approach. As with undirected keys, the first requirement is a morphological database, but a more restricted database than required by undirected keys.
Directed keys are are easier to write than undirected and semi-directed keys because one does not have to create a morphological database first. It is only necessary to divide the taxa remaining at each step into two or a few sets (for dichotomous keys). The major disadvantage is that, if a step relies on a feature not present on a specimen (young buds, flowers, fruits, leaves), then one has to back up a step and follow through both alternatives offered at that step. Whether being forced to understand the words used to describe all the features is a good or bad thing depends on one’s point of view. As a teacher, I prefer directed keys, but they can be frustrating, both for the vocabulary and the requirement for rather complete material. Undirecteded and semidirected keys are easier to use, but they take a lot of work to develop.