Assessing Sophistication

I am on a team of judges for an essay contest. There are around fifty entries in the contest, so I decided to make my selections in a series of cuts. I recently did a first pass over all of them, choosing some to advance to the next round. As I was making the cuts, I wrote down the reasons for the first few decisions I made, then used those criteria to assess the rest. I figured I'd post the criteria here both to get feedback on them and to save them for future retrieval.

I made my decisions primarily using the criterion of general sophistication, but what does that mean? Here's a more specific idea:

• Choice of topic -- while it may be true that some of these topics were suggested by an instructor, I tend to assume that the student has final say in the decision of what to write about, and I chose papers with thoughtful, fresh, unusual topics. While students certainly can do something different with a topic such as underage drinking, it doesn't usually work out that way in practice. I tended to award more cachet to a paper with a topic such as "public education's use of for-profit consulting corporations" than a topic such as "doing drugs while pregnant." By the way, in writing studies, there's a term for that: a "cheerleader paper," an argument no one would possibly argue against. What reasonable person would argue that a woman *should* do illegal recreational drugs while pregnant?
• Command of material -- some of the papers read like they were parroting the investigative style of television news of the 20/20, Today Show, or Dateline NBC variety.* That is to say, they read more like a recital than an argument, they oversimplified the topic, and they repeated banal messages (eat more fruits and vegetables, and do thirty minutes of cardiovascular exercise a day). The papers I chose to go to the next round had more of an authorial presence -- there's an author in these papers who is clearly in control of the subject matter. Some of the papers I chose also used personal experience and/or vivid physical description (pragmatographia) in interesting and effective ways to bring the material to life.
• Marshaling of evidence -- this is related to "command of material." In the papers I chose, there seems to be evidence which is deliberately chosen (not just a data dump) and arranged in order to make specific points. In the marshaling of evidence, these writers seem to have an awareness of the credibility of sources. These papers' writers were also better at integrating source material into their own arguments.
• Looking beyond the obvious -- this is related to "command of material" as well insofar as the stating-the-obvious papers were the more recital-esque ones. The writers of the more sophisticated papers seemed thoughtful enough, for example, to recognize when an argument was facile, when a representation was stereotypical, or why a proposed plan to address an issue wouldn't work. They also demonstrated their understanding and consideration of differing views.
• Bibliography -- the papers I chose tended to cite more, and more credible, sources. By the way, all the entries are research-paper type essays.

* Imitation is fine, actually; I understand the impulse to try on various authoritative voices. In fact, it's a good rhetorical strategy. My disappointment with the 20/20 papers has much to do with mass media discourse, and again, I don't know much about the context -- they could have had teachers who assigned 20/20 in class, for all I know. I contend, however, that more sophisticated readers and writers can recognize the canned quality of that style and try to avoid it.