Statistics are often manipulated to convince people of a certain point of view, or to falsely frighten the public. Statistics might be hardly manipulated by presenting misleading representations, exploiting the tendency to assume that a correlation demonstrates causation, representing irrelevant facts based on what people belief not on what is true, providing anecdotes often emotional in nature and appealing emotion rather than logic. Statistics can mislead if it is not peer-reviewed. It is like a quality control. Without peer review, a report full of erroneous and misleading statistics can be passed off to the public as a scientific report.
An example can be the anti-smoking movement these days that try to convince people through the statistics. Although smoking is hazardous to health but proving it by the means of statistics is not always correct. For instance, Japan has the world's highest smoking population having 67% or so of the population being smokers. It is also the world's healthiest nation. China has the second highest smoking rate and it is the second highest healthiest too.
Some countries still consider statistics as confidential information that should not be disclosed because their enemies could use it to undermine the security and order in the country. We think that this is legitimate but this should not be used as an excuse to hide the statistics that have nothing to do with national security and we think that this matter should be discussed more thoroughly.