Questions about who I should ask a letter of recommendation

<p>Hi, </p>

<pre><code> I'm going to apply for grad school soon and I'm considering who I should ask a letter of recommendation.
I've already had a professor who I'm currently doing research with and I've also taken one of his classes. So he should be perfect to write me a letter.
The second I think will be from the professor who I've taken three classes with.
I'm debating on the third letter. An instructor from my department used to supervise the software development team I was in, I'm quite familiar with him and i'm also taking his class this semester. I can ask him to write me a letter. But the problem is (maybe not a problem...), I think letters of recommendation should be from professors? At least with a PhD degree, but he doesn't have a PhD...will the recommendation letter from him have a lot of "weight"..or is convincing..? Beside the instructor, I can also ask the other professor for a letter instead, who I took two classes with and got good grades.

Please give me some advice on this, in this case should your references always be someone with PhD?
</code></pre>

<p>Thank you very much</p>

<p>The letter doesn’t need to be from a professor that has a Ph.D; it does need to be from someone who is articulate. Make sure you did well in their class–B or higher. If they don’t sound enthusiastic about writing you a letter, look elsewhere.</p>

<p>I recommend getting a letter from someone who is well published and well respected in a field that you are trying to get into for graduate school. For instance, I worked for a professor last summer that is famous for signal processing and machine learning and is very well known (has done TED talks for instance). A recommendation from him will carry a lot of weight in signal processing and machine learning, and some weight in electrical engineering in general.</p>

<p>er…still hard to decide…thank you though</p>

<p>I got my letters from my PI (PhD, DVM), a postdoc in the lab, and a grad student that I had worked closely with in a previous lab. They seemed to work well.</p>

<p>Follow this advice:</p>

<p><a href=“http://www.cs.cmu.edu/~harchol/gradschooltalk.pdf[/url]”>http://www.cs.cmu.edu/~harchol/gradschooltalk.pdf&lt;/a&gt;&lt;/p&gt;

<p>It’s rather good. Goodluck</p>