Unless you work with metadata regularly, the different definitions of these ideas are confusing. Here are some basic definitions to clarify the terminology. Definitions taken from: Pomerantz, J. (2015). Metadata. MIT Press.
Metadata: "metadata is a statement about a potentially informative object" (Pomerantz, 26). This is a bit more of a nuanced definition than the commonly used definition "data about data".
Metadata Schema: "a set of rules according to which a language operates" (Pomerantz, 30). Example: Dublin Core metadata schema
Syntax Encoding Schema: a set of rules that dictate how to represent, or encode, a specific type of data at the individual level (Pomerantz). This changes how you structure the data even if the data itself isn't changing. Example: ISO 8601 - standard for encoding dates in a standard way.
Controlled Vocabulary: a set of rules that dictate how to represent a specific type of data, at the individual metadata level (Pomerantz). While syntax encoding cares about how a string is formatted (but not what is in the string), a controlled vocabulary provides a finite set of strings to be used. Example: RxNorm - a terminology used to normalize names for clinical drugs and link its names to common drug vocabularies in use.
Ontology: Information science defines ontology as "a formal representation of the universe of things that exist in a specific domain" (Pomerantz, 46). In the context of research data, we use the information science definition of ontology and not the philosophical definition. Example: Oral Health and Disease Ontology - use for representing the diagnosis and treatment of dental maladies.
