{"id":5504,"date":"2026-10-16T16:47:14","date_gmt":"2026-10-16T14:47:14","guid":{"rendered":"https:\/\/www.psyctc.org\/psyctc\/?post_type=docs&p=5504"},"modified":"2026-10-17T14:38:13","modified_gmt":"2026-10-17T12:38:13","password":"","slug":"hashing-and-hash-codes","status":"publish","type":"docs","link":"https:\/\/www.psyctc.org\/psyctc\/glossary2\/hashing-and-hash-codes\/","title":{"rendered":"Hashing and hash codes"},"content":{"rendered":"\n

Nothing to do with food. (I have just discovered that it comes from the lay meaning of hashing: to chop up or make a mess, see the history part<\/a> of an excellent Wikipedia article.) <\/p>\n\n\n\n

Details<\/h4>\n\n\n\n

Hashing is a one way conversion of a set of data to a set of data that looks like gibberish but which preserves a 1:1 relationship with the hashed data. Suppose you had data with participants’ names like this.<\/p>\n\n\n\n\n\n\n\t\n\t\n\t\n\t\n\t\n\t
\"Chris Evans\"<\/td>\n<\/tr>\n
\"John Smith\"<\/td>\n<\/tr>\n
\"Jane Doe\"<\/td>\n<\/tr>\n
\"Cyril Brown\"<\/td>\n<\/tr>\n
\"Chris Evans\"<\/td>\n<\/tr>\n
\"Jane Doe\"<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n\n

Then hash codes for those names might be like this.<\/p>\n\n\n\n\n\n\n\t\n\n\t\n\t\n\t\n\t\n\t\n\t
V1<\/th>hashVal<\/th>\n<\/tr>\n<\/thead>\n
\"Chris Evans\"<\/td>61be1b8203762e2e44125bf2cc349f0be8396ccdc6c4459805bc5190bdd38b1f<\/td>\n<\/tr>\n
\"John Smith\"<\/td>37def76d8e1831e6fd5b8242a23a63ffda5d4ff564055239b90f665f7b288d91<\/td>\n<\/tr>\n
\"Jane Doe\"<\/td>f92677b38e01d005619d8c03339858767f6d665ad98473f2a9d1939cb78d4e92<\/td>\n<\/tr>\n
\"Cyril Brown\"<\/td>8289e99db0654909336f396c76da65e16cfaae3908e7f407816a7edee9ce80b9<\/td>\n<\/tr>\n
\"Chris Evans\"<\/td>61be1b8203762e2e44125bf2cc349f0be8396ccdc6c4459805bc5190bdd38b1f<\/td>\n<\/tr>\n
\"Jane Doe\"<\/td>f92677b38e01d005619d8c03339858767f6d665ad98473f2a9d1939cb78d4e92<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\n\n\n\n

That’s a tiny example but what you can see is that the hash codes for the two repeated names “Chris Evans” and “Jane Doe” remain the same and you can probably see that each distinct name got its own hashed value. This makes it easy to pseudonymise your data: you are not completely anonymising it, which would throw away the ability to know that the data from “Chris Evans” are from the rows with that name (and likewise, for “Jane Doe”) and hashing also ensures that the distinct names get distinct hash codes. <\/p>\n\n\n\n

There are various algorithms to create hash codes (those were generated using the sha512 function from the openssl R package<\/a> and all of them require that you give the program a key, a short word or phrase (I used “sausages”). You can do this for your own data using my shiny package here<\/a>. What all the algorithms share is that you can’t get back from the hashed values to the original values. Different algorithms can handle larger numbers of unique values to hash without eventually giving two original values the same hashed value.<\/p>\n\n\n\n

Try also<\/h4>\n\n\n\n