Scientists determine the probability of collecting their data for an experiment by chance alone. The p-value for a given experiment is an indication of how likely that data would be obtained if the independent variable had absolutely NO effect. Knowing this, explain why one cannot prove there is a difference between treatment groups and explain how our "confidence" that a real mean difference exists changes as the p-value gets smaller.