I'd like to get some help with a question on Likert scales. I'm doing some work on a study that uses Likert scales and have been needing to know if the data that come out of them are ordinal.

My lecturer has told me that data coming out of Likert scales are nominal but none of the literature I've looked at backs this up. I've challenged him on this and he says that there is some debate as to whether data from Likert scales are nominal or ordinal. Again I can't find anything to support this. I've found literature on the debate as to whether data are ordinal or interval, but nothing on data being nominal.

Could I get people's ideas on this? I'm not statistically minded and this has got me confused.