A widespread assumption among UX professionals says that users are not keen to use the virtual keyboard. Guidelines for mobile UX recommend to minimise text entry, suggest that typing holds back the mobile experience and recommend to use selection controls over text entry fields. Are there any studies that support (or contradict) this claim?
The problems with mobile keyboards are far beyond "virtual" and there's more reason to minimize their use than "people don't like them". For a good overview of the topic read Pickup Usability Dominates: a brief history of mobile text entry research and adoption. There's a wide variety of HCI research on mobile text input, from physical keyboards to virtual, QWERTY to non-standard layouts. This research doesn't exist just because people are curious; mobile text input is a serious problem. There's also plenty of other research out there; just go to Google Scholar and search "mobile text entry HCI" or something similar.
Mobile keyboards are small, your fingers are (relatively) large. You're using the device while walking. Correcting mistakes is more difficult too. Mobile text entry is sub-optimal at best.
A telling statistic from the above article is that "expert" soft keyboard users type significantly slower than physical keyboard typists:
Their results estimate that an expert user could achieve 40+ wpm on a soft QWERTY keyboard with novice soft-keyboard users achieving around 20 wpm (MacKenzie, Zhang, & Soukoreff, 1999).
IN contrast, average professional keyboard typists get around 50 WPM and experts can be as high as 120 WPM, though interestingly novice use of physical keyboards is close to the same 20 WPM.
A common problem for research in this area is it's often conducted in a lab (the above study sites a meta-study finding about 70% of tests were in laboratories, 20% in the field and 10% surveys). Mobile usage is often, well, mobile, meaning you could be outside, where it's hard to see your backlit screen in the sun, where there's all sorts of noise around you and you might even be walking as you type. All of these contribute to a higher error rate.
Now there are plenty of mobile keyboards that attempt to improve WPM and accuracy, but they always come at a massive cost to first-time usability. This is pretty much why QWERTY dominates the smartphone world (and the physical keyboard world as a matter of fact)
Despite its suboptimality and problems on small devices, both market trends and some user tests point to the QWERTY keyboard taking on this role, either as a physical or an on-screen keyboard.
Mobile keyboards slow people down. Mobile keyboards introduce errors. People hate being slowed down and making errors; you don't need research to know that. Even if people loved using keyboards that slowed them down and introduced tons of errors into their work, you should still avoid giving them to users as often as possible.
Is it common knowledge? No. I don't think so.
Given that we still see most text entry (SMS, email, etc.) is still by keyboard even though voice to text exists on new phones - users may not rate "Using Keyboard" as the most fun thing to do with their new phones - but they often prefer it.
But - I think this is too broad a question to answer well. It comes down to the app, what it gives the user and the overall experience - not just the input type.
And with all things... ask the users directly - test... test... and test again. Personally, I find "common knowledge" in USER experience to be an oxymoron. If it existed we wouldn't need to test.