If you capture information from users about institutions, you can expect to see a lot of variation. Various names will be used by different people depending on their own biases e.g. first language, view of the institutional structure, preferences for short names or acronyms, etc.
If you manage any system that captures this sort of information, you have probably already noticed how difficult it is to report consistently on your data because of name variety.
Use GRID to capture ids rather than labels so your data is always consistent. If institutional names and details change over time, it will not affect you because GRID ids are persistent and their metadata (such as name and location) will be updated. By using standard tools and libraries, such as autocomplete and typeahead, GRID can help solve your data capture woes.