I find it discouraging that we can only seem to discuss GPS vs Traditional methods as a binary either/or question.
What I see lacking in GPS dependent pilots is not so much a GPS issue as a more general lack of TLAR (That Looks About Right) estimating ability.
What I want to hear from my students is an approximation of the heading, distance and ETE before they hit the enter button. Developing that ability gives you the skills to realize when the box is not telling you what you want and keeps your SA up if you loose the GPS.
Most real world GPS F*uck ups involve a gross navigational error caused by incorrect data entry. So if you went to Lower Gum Boot Upon Tyne for a bacon sarny and the heading to get their was 310, if the box says the heading to get back home is 223, then you should immediately shout "Hey that ain't Right !"
I think it is also important to note that "traditional" navigation techniques were developed because they were the best techniques for the technology that existed at the time. The principal technological deficit then was that there was no way to know exactly where you were in real time and thus traditional nav techniques were developed to overcome this limitation.
Well technology has moved on and now GPS provides real time extremely accurate positioning data so the principal factor underpinning traditional nav no longer applies. I don't have a lot if time for the Luddites that reject GPS. It is a better way to navigate in every way. Using GPS makes your flight safer and more efficient and greatly reduces the chance of inadvertent penetration of the modern thickets of controlled/restricted airspace.
When teaching nav I start with traditional nav because it teaches the fundamentals of course/heading/time/speed/distance without which having a good TLAR sense is impossible. But when that is mastered I am clear on how, when you want to get from A to B in the real world, you use the best information available and that will always be provided by a GPS.