Does your field require statistical analysis?

I use some basic statistics (student's T, means and standard deviations, Z-score normalisation, etc.) in my acoustics, but it's not earth shattering if I get a false positive. There's a bit of fudging with population - because of the data, I shouldn't be using some of the means due to low populations. Oh, well.

Sure, we use lots of statistics... in Biostatistics. ;)

Let's see, nuggets of advice when reporting statistical results.... A common standard is that your statistical write-up should be detailed enough so that a knowledgeable reader could replicate your results if given your data. That's probably a good place to start, and then add further elaboration if you think your methods need justification or your audience won't be able to correctly interpret the results unaided.

I understand your pain: I'm trying to deal with adding the appropriate statistics (understandable to epidemiologists) to a field that doesn't have standard stats applications (risk analysis and Monte Carlo simulation). We have great debates in my modeling group about what statistics are useful . . . and whether reviewers will accept them. The consensus seems to be that, as johnascott said, you need to explain your stats so they can be replicated by someone within your own field. That means they should be understood by your colleagues. If you're doing stuff that people in your field can't understand, you might want to consider publishing on your methods first, then your results. Sounds like that would be your best bet, so I suggest you make friends with a statistician (that's what my advisor did with a similar problem)!

I am in the experimental part of Psychology- and I must say that stats pretty much rule everything. I use a program however- SPSS and that helps loads.

Of course, the old saying that "statistics can be made to say anything" is mostly true, which makes it difficult to analyize them in the most productive way.