Critiquing 101: So What DO I do?

I do a lot of critiquing/beta reading. Mostly through, but I also have participated in real life critique groups from time to time, and I'll agree to read other writers' work fairly frequently (this is not me asking you to critique your novels. Chances are, I'm already way behind on agreed upon reading right now).
Getting input on your writing is so important, but so many writers are hesitant to give constructive criticism, or don't know how, or don't feel qualified.
In this series, I plan to discuss pitfalls to avoid as well as things you can do to make your advice easier to understand and more helpful to the writer.

Some subjects we'll be covering:

If you've read my previous posts, you may have noticed a lot of them feature what 'not to do' rather than what to actually do. 
I've given some 'to do' advice here and there (be specific, include positives as well as negatives, explain why you make suggestions, use common sense), but here I'm going to go into more depth. What exactly do you say in a critique?

Give your impressions: BOTH positive and negative. A lot of critiquers say they don't bother stating the positive because it won't help the writer improve. You don't get critiques to tell you how good you are. You need them to find out how to be better. To some extent the reasoning is correct, but the conclusion is faulty. It is important to know what is working, especially because sometimes input will conflict. If you don't mention something is your favorite part, it may end up being another critiquer's least favorite part, and your favorite part may end up on the cutting board during edits. 
Humor is also incredibly subjective, so mentioning if something is funny helps (with the side effect of if something wasn't meant to be funny, they'll probably want to know that too). 

Mention Any Confusion: If you are just staring at a passage thinking, "I have no idea what is going on here?" definitely say something. However, highlighting large sections and just saying, "I was confused through here" can add to confusion on the writer's side. I've had readers do this to me, and I just blink at the section thinking, "It's pretty dang clear. How is this confusing?"
Try to break down exactly why you are confused. Maybe parrot back what you think is happening, even if you aren't sure it is right. Point out any sentences that were particularly problematic: if a sentence has two possible meanings and you weren't sure which to choose, for instance, or if a character's choice doesn't make sense to you or you aren't sure why he said something. Generally the more specific the better.

Suggest small changes: Does a word just not fit right to you? Suggest a better word. 

Read the work out loud: This is great advice for self-editing, too, but if something trips your tongue when read aloud, it probably could be smoothed out. Point it out.

Point out repeated phrasing/ideas/actions: Repetition has its place, but a lot of authors tend to have pet words and phrases they overuse or fall back on when they are out of ideas. And most of them won't even be aware of it. Ask if it is intentional. "You know your main character has rolled his eyes three times in this scene. Intentional?" Point out if the writer tells you something you already know. "You have it tell me about how the political system works here, but all this information was already given in the last chapter, and I remember it." 

Point out inconsistencies: Is the character's hair black in one chapter and blond in the next? Does she mention being a vegetarian and then order a BLT? If you want to be really good, try writing down small details or keeping a notepad document open while you read on your computer and tracking how they change. Timing can be difficult for a lot of authors too. Like you say something happened three days ago, but the next day it was suddenly a week ago, like we skipped ahead several days. This especially can get confusing if the author has done edits and maybe missed a few spots that should've been changed to incorporate the new writing. 

So this is the end of my series on Critiquing. I don't consider this an extensive "how to" by any means, more of a list of observations about what works and doesn't work. I may add onto it and tweak it in the future, but for now, I just hope this was helpful to you in some way.