For my analysis, I chose an article about the leadership of the Free Syrian Army moving its base to Syria from Turkey.
The first attribution takes place in the first paragraph. The author attributes the fact that they are moving to something their leader said Saturday. It says "the leader announced" which would make it seem as though they aren't getting the information firsthand, which is true. The very beginning of the second paragraph talks about a video the leader posted on the internet, which sets up the knowledge that most of the information in the article will come from the video. I believe this is effective, because now the reader knows where the information is coming from, and isn't fooled into thinking that anyone at CNN actually sat down and did an interview with him.
The next source cited is a defense analyst for the Washington Institute of Near East Policy. The name a specific person, but again in the paragraph following this citation, we find out that the story is quoting an essay that was posted in August. This is a credible source, because it's someone who works at an official government agency, so we can trust what he says. However, I don't like that they use an essay posted in August, because the situation has changed a lot since then, and the essay doesn't talk about them moving from Turkey to Syria. I think that it would have been more effective to get a quote from this person, or someone else in the agency about the current situation, and how it will affect not only that region, but things in Washington as well.
The rest of the article is just a recap of events that have happened recently in Syria, and again, there is plenty of credible attribution. And as with the previous examples, all the attributions are taken from command statements posted in the past. This works much better, because it's just a recap, so we don't need a quote from somebody that is current, because those situations are no longer current.