A few comments:
The first thing we need to consider is the type of pain being assessed. What Pesta (Aside to Pesta: SDT really isn't appropriate for pain. See Rollman in the mid 1980s. Exact ref available when I get home) Mercutio and possibly others are describing is experimentally applied painful stimulation - usually (but not always) applied to healthy volunteers. Interesting stuff, but the question remains as to whether or not the results found are applicable to clinically relevant pain (In fact, I just accepted an invitation to write a paper on exactly this topic).
The biggest issue is that the experimental situation simply cannot parallel the clinical situation. Pain is a subjective multidimensional experience which is influenced by biological, cognitive, and emotional factors including things like a person's age, gender, previous experience, mood, fear of pain, and most important to this discussion, the meaning of the situation. There is no way that the pain I apply to a subject in a carefully controlled experiment where he/she knows they will not have any permanent damage and can terminate the stimulation at any time can parallel the pain felt by a cancer patient who is afraid that the pain will be unending and may signal disease progression etc. So, many of the experimental methods are not really clinically relevant.
Clinically, the most common pain measure is the numeric rating of 0-10 which has been endorsed for clinical trials and other types of studies and across the lifespan. Although it has some problems psychometrically, our understanding of its properties is continually growing. For instance, there is ongoing work into clinically meaningful differences etc. And, it is easier for more vulnerable people - like children and the elderly - to use than visual analog scales. Of course, it can't be used for people who cannot express their pain verbally. But many groups (mine included) are working to develop methods to assess pain in people who cannot communicate verbally.
Up to this point, we've only talked about the intensity of pain or how much it hurts. Also important are pain qualities, or how the pain feels. The most common measure of this is the McGill Pain Questionnaire which is a list of adjectives describing the sensory, affective and evaluative qualities of pain. So, this measure helps us to discriminate different types of pain: for instance, a tension headache, migraine and a neuropathy might all be rated 7/10 but the tension headache might feel "dull" and "heavy", the migraine "pounding" and the neuropathy "burning" "shocklike". Those words are really helpful with diagnosis and with understanding the effects of different treatments.
Finally, a good pain assessment includes a measure of pain's impact on physical, psychological and social function. Usually, this is measured by asking to what extent the pain intereferes with desired activity, mood etc.
As for comparing across different types of pain. Extremely difficult to do, but has been done with the MPQ. I once posted a chart here of the rankings of different types of pain. I'm not on my computer right now so can't find it at this moment. I'll try to search for it later.
Gender and pain: huge issue. The best answer to this one is that the differences are the final outcome of psychological, social, and biological differences between the sexes. Hormones definitely play a role (and Jeff Mogil has been showing that female rats may have a unique endogenous pain inhibitory system). However, in humans, boys and girls are taught different things about the acceptability of expressing pain, and we know that men and women employ slightly different coping styles and are responded to differently by the medical system. So, the gender differences are multidetermined. None of the gender and pain researchers I work with think there is a unifactorial explanation.