[greymule]

In logic, symbols have clear, agreed-upon meanings, and the student starts from a platform of uniformly defined terms: equals, greater than, less than, not equal to, some, all, if, only if, and so on. Every statement can be proved to be either true or false. All the analysis goes into determining truth value, so tests in logic pretty much have to be true/false in one form or another.

One problem with true/false tests in sociology, history, psychology, baseball, and so on, is that ambiguity (which is not found in logic, as least as far as I remember) involving a single word can affect the perceived truth or falseness of the entire statement. Look at the problem we had with "first base after an award" versus "next base after an award." T/F tests can sometimes be useful, but they have to be constructed very intelligently.

I'm reminded of a quasi T/F questionnaire my Congressman sent me not long ago. He listed a bunch of statements, and I was supposed to check either the "agree" or the "disagree" box next to it. I think there was also a "no opinion/not sure" box, too. One of the statements was "I would like to see the U.S. resume trade with Cuba." Anyone can see where my Congressman was coming from, but to me the answer to that question requires an explanation and not a true/false answer. Other statements were similar: "The government should do more to help the public schools," "reasonable firearms laws would help keep guns out of the hands of schoolchildren," "the rich should pay their fair share of taxes." To answer these questions, you have to do something familiar to most people who take a lot of tests: try to read the mind of the questioner and answer accordingly.

PS. I did not check any boxes but instead wrote out answers. I was amazed when I received a response indicating that somebody had actually read what I wrote!