The following is from Learn clojure in Y minutes.
array maps and hash maps share the map interface.
array maps - less fast lookup, retains key order
hash maps - fast lookup, doesn't retain key order
> (class {:a 1 :b 2 :c 3})
clojure.lang.PersistentArrayMap
> (class (hash-map :a 1 :b 2 :c 3))
clojure.lang.PersistentHashMap
Array maps eventually become hash maps after they get sufficiently large.
keyword - like strings, but with efficiency bonuses
Maps can use any hashable type as a key, but usually keywords are best.
> (class :a)
clojure.lang.Keyword
> (def stringmap {"a" 1 "b" 2 "c" 3})
#'learn-clojure.core/stringmap
> (def keymap {:a 1 :b 2 :c 3})
#'learn-clojure.core/keymap
Note: Commas in a map are treated the same as whitespace.
You can retrieve a value from a map by calling it as a function.
> (stringmap "a")
1
> (keymap :b)
2
This operation is commutative with keywords, but not with strings.
> (:b keymap)
2
Note: A non-existent key in a map yields nil.
assoc - used to add new keys to hash maps
dissoc - used to remove keys from hash maps
> (def newkeymap (assoc keymap :d 4))
#'learn-clojure.core/newkeymap
> newkeymap
{:c 3, :b 2, :d 4, :a 1}
> (dissoc keymap :b)
{:b 2, :a 1}
Don't forget that clojure types are immutable.
> keymap
{:c 3, :b 2, :a 1}