A statistic is not a number! Unfortunately in everyday language, people often use the words "statistics" and "numbers"interchangably. I think part of the blame may rest with the extensive use of numbers and statistics in sports. As of May 31, Baltimore Oriole Adam Jones had a batting average of 0.228, the statistic is the batting average; 0.228 is the computed value for the specific batting average for Adam Jones.
Remember, a statistic is a function, a procedure used on sample data in order to estimate a population parameter. Thebatting average is the statistic and it is the following procedure: count the total number of times a player goes to bat and count the number of times the player gets a hit. (This is oversimplified so as not to confuse non-baseball fans but the actual procedure is very detailed and specific. If the player "gets a walk" that at bat is not counted, etc. etc. etc.) Compute the batting average as: #hits/#atbats, a batting average is a kind of proportion. I would describe the parameter it is trying to estimate is the underlying population probability (Bernoulli distribution) that a specific player will get a hit at one "at bat."
Please discuss an example where you describe a statistic (procedure) and the parameter it is trying to estimate. As is the case with most statistics, assume that the parameter is a value which we will never truly "know." So, for example, I might say that to truly know the population probability for Adam Jones I would have to observe him batting one million times (poor guy, that would be *a lot* of batting).