Zooarchaeologists often consider fragment counts as essential primary data.  For English language users this is “NISP” (Number of Identified Specimens). In recent years this has been contrasted by some with “NSP” (Number of Specimens).  The intent often seems to be to supply an abbreviated label, by way of acronym, for the quantity of fragments that have identified at a less specific taxonomic level. However, because the level of taxonomic specificity used in any particular study varies with research goals, preservation, available comparatives, time, and funding, the boundary between what belongs in NISP and what belongs in NSP is not standardized. This renders the use of NSP problematic, particularly in any attempts to compare data between studies.

A popular alternative has long been to provide a primary data table that lists all the levels of taxonomic identification and size category identification used in a particular study, thus communicating directly the count of fragments at each named level. All these counts would be considered NISP.  All represent identification at some level. The term “undifferentiated” can be inserted as a qualifier for a subset of any named category, such as “undifferentiated mammal” or “undifferentiated vertebrate”. This approach has much to recommend it. It makes the type and level of identification transparent, and links the count to it directly. At allows quick facility in aggregating the tabled data in multiple alternative ways while maintaining clarity about sample size for that particular analysis or comparison.