Identifying a scholarly article is easy enough. Once you've seen a few scholarly articles, it's easy to tell them apart from magazine and newspaper articles. Most databases even have options (that work pretty well) to filter out everything but scholarly articles.
What's harder is knowing whether a scholarly article is any good. Many times you don't even need to know. A scholarly article, any scholarly article, will be "good enough." But if you really want to impress your readers (including the professor) or get the best info it's worth taking a minute to ask more than "is this scholarly?"
Below are some questions you can ask yourself when you want to weigh the quality of a scholarly article.
One way to determine just how good a scholarly article is by looking at who wrote it.
Scholarly articles always include the name and credentials of the author. Credentials include details like the author's level of education and the place they work.
Start with where the author works. Right or wrong, an article will carry more weight if it is written by a professor at a top university such as Harvard or MIT. If you've never heard of the college or university that the author works for don't sweat it.
Now look at the author's name. Do they have a PhD or some other advanced degree? Is it a name you've heard of? It probably won't be, and that's fine, but then again it just might be one of the "big names" mentioned in your class readings.
If the author is one you've never heard of, don't be afraid of Googling them. Many academics will have a CV (fancy resume) that is easily available online. This will tell you how much they've written before.
There are a few prestigious scholarly journals that are somewhat widely known: The New England Journal of Medicine, Science, Nature, and a few others. These are the ones that get mentioned in the news from time to time. In addition to these, there are many thousands of journals you've probably never heard of: Antipode, Macroeconomic Dynamics, Canadian Issues, Library Hi Tech, etc. The library subscribes to dozens and dozens of scholarly journals for most academic disciplines.
To help academics wade through the increasing number of journals, several ranking methods have been developed. These are mostly based on different kinds of automated measurements, such as the number of times a journal's articles are cited in other articles. You can read about the different measurements here.
There are a few websites that publish, for free, different journal rankings:
Finally, do read the article. Don't cite an article just because it was written by a famous scholar for a prestigious journal. Cite it because you actually used it in your paper; and use it in your paper because it contributes to the point you're trying to make.