XML can be a bit tricky to set up correctly, but here's a detailed explanation of an near real world example:
Let's say you this is the response you receive:
<?xml version="1.0" encoding="utf-8"?>
<collegepostresponse xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns="http://www.collegeinfousa.com/providers/">
This is the correct configuration with the treat response as XML checkbox selected:
Success POST String: true
XML Response XPath: //x:collegepostresponse/x:accepted
Here's a fairly technical explanation of why we do it this way:
The trick with this xml response is that it is overriding the default xml namespace. As a result, you can't just use the // and / XPath operators. So as a workaround we add an alias "x" for the namespace. The result is the appropriate xpath for the "accepted" tag is "//x:collegepostresponse/x:accepted"
In the collegepostresponse tag, you see all of the namespaces. The named ones are "xmlns:xsi" and "xmlns:xsd", which can be referenced via XPath as "xsi:" and "xsd:". The one without a name overrides the normal behaviour for unnamed path references, and it leaves no way to refer to the correct thing. So, the "x:" alias allows you to refer to the overridden namespace.