Look at it from the mindset of the reader.
By the time I've read the 5th personal statement they start to meld into each other. Also, like movies, pretty much every personal statement says an often repeated theme. If you're going to make your personal statement very long it's just going to make that theme that is with over 99.9% likelihood nothing new to the reader that much more boring. (E.g. my brother/sister/girlfriend/mother/father, etc has depression/schizophrenia/bipolar disorder/anxiety and this makes me extremely impassioned/enthusiastic/motivated to treat mental health. To quote Abraham Lincoln/George Washington/Hippocrates/Sigmund Freud/Albert Schweizter..........
Personal statements are only going to stand out if there's something exceptional (and maybe not in a good way) about them just like in a crowd of 100 black sheep the white one stands out. E.g. To quote the highly respected lover of the human form and expert in experiential pleasure Hugh Hefner (okay now you got my attention, but now this now exceptional letter is likely heading for a nosedive).
However, I have also heard that it should be longer for psych.
I don't know where that came from and find no reason to believe it has any merit. I've said this before in other threads. In industrial psychology studies letters of recommendations and personal statements pretty much have no bearing in having any metric in predicting if the candidate will be a good fit for a position.
For the above reason I sometimes find it odd how much emphasis is put on these things, and find it odd that no program I've seen ever looked into the science of the application process itself that has been thoroughly studied in the field of psychology. For a field to pride itself on evidenced-based data, then not use such data or employ highly trained experts for their advice on such matters (which is what patients do with their physicians, ask us because we're the experts) then why aren't we doing it ourselves?
Should LORs and PSs be thrown out? No. It's one of the only methods the program can get to you know you as an individual. Despite what I said above I do think on rare occasions the LOR and/or the PS could've made a defining substantive difference but this is only with extremely exceptional material. E.g. if Phil Resnick, one of the greatest forensic psychiatrists ever wrote a very impressive LOR, and he is known not to laud anyone unless that person was very good, I'd pretty much let that person into a program almost based on that 1 letter alone. Again this is very rare and not to be expected.
And aside from the personal characteristics of the writer, it gives an opportunity for the program to judge a person on their writing and this is something important in medicine. A longer PS just gives a program more ammo to find a flaw. This is another reason to not make your PS longer than needed. You're just possibly giving yourself more rope for the hanging if you don't write well. If you make it long it better be damned worth it.
