Saw an article about surname-changing the other day. I honestly don't remember whether the author was justifying keeping her maiden name or explaining why she didn't, but it got me thinking about the subject.
When I was a grad student, there was an incident in which a postdoctoral fellow's daycare provider repeatedly called our lab trying to find her, and those of us answering the phone kept telling the caller that she had the wrong number because we didn't recognize the name she was saying. Turned out that this postdoc (who was new to the lab at the time) had two children whose surname was the same as her husband's. We knew her as Janet X, so when someone called the lab asking for "Mrs. Y," we just didn't make the connection. Fortunately, we eventually figured it out and things were fine. But it did bring home for me one complication of not all having the same family name.
In science, your surname is everything, since your publication record (books and articles published in scientific journals) is what gets you your doctorate, your job, and the grant money without which you have no functioning lab. There are strictly observed conventions for how authors are listed in these publications, too: the first author listed did most of the work, and the last author is the most important person on the list, usually the person in whose lab the work was done. (Whose name is listed where in the order of authors is a big deal and often the subject of big arguments!) Accordingly, women almost never change their surname after establishing a publication record, because then it's harder for people to keep track of what work you've done. The usual change you see is hyphenation, if there is any change at all.
For a variety of reasons, I decided early on in the graduate school process that I had zero desire to be an academic professor. I like and respect the professor in whose lab I did my graduate work, but I saw what his life was like and wanted no part of it. With the help of an internship that this professor (my thesis advisor) was kind enough to let me pursue on his lab time, I was able to transition into a different career right out of grad school, this career not requiring a publication record, and also having much more job security than academia.
Himself and I were married about two years after I finished grad school. Since by that point I had not only a publication record but two years of history at a company, I decided to keep my maiden name as my middle name and take my husband's surname, figuring that anyone who needed to find me would then be able to find me. A good idea in theory, but a major nightmare in practice!
First stop: Social Security office, marriage license in hand. No problem there, actually. New middle initial, new last name.
Then, the DMV, who in their smug, bureaucratic way, informed me that my marriage only changed my surname, not my middle name. They issued me a new driver's license with my new surname and old middle initial. A week later, I went back and showed them my new Social Security card and innocently said that there must have been a mistake with the middle initial on my new driver's license. 15 min later, new driver's license with correct (new) middle initial. Ha! Take that, bureaucrats.
Last: a new passport. For any who have not had the fun of getting this far into the process, if your current passport is still valid, they don't issue you a new one. They just put an addendum into the back with your new name, so until it expires, you have a passport with two different surnames in it.
Was I done there? Sadly, no. At my company, I had to provide notice of my name change three separate times...to Payroll, the guy who controlled the phone list, and then the guy who assigned email addresses. Nuts.
This whole process took me several months to complete. A work friend who married a few months after I did decided to just keep her maiden name after watching the hassle I went through.
I love my husband dearly, which is a good thing since I'm never changing my name again!