How many dependent variables should be in an experiment?

Ideally, one.

When you do an experiment, you're basically finding the following:

"When I do (some independent variable), then (some dependent variable) will happen." An example: "If I were to grab Michelle Obama's purse, then she would yell at me."

Of course, there can be more than one dependent variable in an experiment. For the example I gave, my rude action would result in:

• Me getting yelled at
• Me getting wrestled to the ground by the Secret Service
• Me going to jail
• Me going to court for being an idiot
• Me getting ridiculed by the national media

All of which are dependent variables in this case.

For scientific purposes, it's usually best to change only one thing (the independent variable) and look for one resulting change (the dependent variable).

Good exposition of dependent and independent variables with additional info here http://www.cool-science-projects.com/independent-and-dependent-variables.html