[thakitt]

Do what all normal researchers do.
Put those demographic questions at the end so they've already finished the important part before they get pissed. Use whatever terms you need to do your analysis but understand that not everyone is going to cooperate. There is no other way. Unless you can define questions that can get that information without them knowing, which is usually not effective and can make your results not reliable.

[fedordzen]

Go with what rah said.

[antiggill]

Is the "Other" option for the untouchables? What is the current PC term for the untouchables, by the way?

[Speareerfug]

You should also make it clear that the answers will only be used as grouped data; basically that the information is "blind" and will never be personally identifiable.

(if that is, in fact, the case)

[Averti$ingGuru]

cultural sensitivity at it's best

[forexsoft]

End the questionnaire with "Are you or your family from a subhuman stock not worthy of being pissed on by regular folk?" I'm certain you'll get a good response.

[kvitacencia]

First contact your IRB to see whether you can ask such questions for your study.

What's an IRB? This isn't a study which is sponsored by any research body, I'm just doing it for my own curiosity, and the primary reason is to find out about reading habits in general - what the distribution looks like.

Could including an explanatory note before that section, telling the candidate how there is no neutral language, and thus why such terminology is necessary, work?

[Ervntewc]

Rah has this spot on.

[tobaccoman]

It's generally either a mark of Shaivism/Vaishnavism, or a caste mark. This is WRT to those from the south. If it's someone else, then it's probably vibhuti or angara, meaning "sacred ash".

I had to google it, but I think I understand now. But how do I recognise a southern Indian? (Btw, does that mean a Dravidic person, not an Indo-Aryan?) Can a name help?
Some Indians here have "stereotypical" long names with easily identifiable roots, like Rajasekhar or Mahalingam . Some only have a short name with initials, like P J Harvey (name changed to protect the innocent). Maybe their accents or skin tone can be of any use?

[angelxmagic]

I had to google it, but I think I understand now. But how do I recognise a southern Indian? (Btw, does that mean a Dravidic person, not an Indo-Aryan?) Can a name help?

Generally, yes, south Indian means of Dravidian extraction. Southern names are quite often distinctive, as you pointed out below.

Some Indians here have "stereotypical" long names with easily identifiable roots, like Rajasekhar or Mahalingam .

Those are the stereotypically South Indian names that I'm talking about, but you can never be sure.

Some only have a short name with initials, like P J Harvey (name changed to protect the innocent). Maybe their accents or skin tone can be of any use?

The accent is something which helps me distinguish, but it can be hard to know for someone who doesn't know the nuances of linguistic variation in India. For instance, the difference between a Punjabi and a Tamil accent is glaringly obvious to someone from like me, from Maharashtra, but it may not be to someone who isn't familiar with them.

The skin tone is a definite indicator, but the problem is that you'll find that it correlates quite often to caste and economic background as much as position on the North-South axis. As a general rule, the more you move towards the South, the darker the skin gets.

[SpeavaJap]

One Indian we had had a very peculiar accent. Final "t"s in all his words were replaced with a popping sound, the one soap bubbles make in cartoons.
What do North Indian names look like?

[Caursedus]

Agree with Rah. Put the caste question as this:

"which caste do you believe best represents you?"

And include the untouchable, the proper term. It pays to be as accurate as possible with the options, and as vague as possible with the questions.

[djmassk]

And you want to go to top school in the US? This entire thread is a joke, right?

You are aware that:

a) This is not my field, and
b) This is not a sponsored or formal study, and
c) I'm doing this purely for my curiosity, and finally
d) I'm not trained in any of this, not being an experimentalist but an engineer,

right?

[avaiptutt]

Yes, that was all factored into my assessment.

I'd like to know more. Could you point out the problems that you see?