Just because you find a source of data or statistics, that doesn't mean it is necessarily useful or reliable. Some factors to consider:

• Who collected the data or produced the statistics, and why? Also, who funded the collecting/producing, and why? Is the source reliable? For example, an advocacy group has an interest in accuracy to accomplish their goals, but they also have an interest in the outcome and so may (intentionally or not) slant their collection of data or presentation of statistics.
• How were the data collected?
  • Did they count, measure, estimate? Were the things counted significant, or just easy to count? How were measurements or estimates made?
  • Complete coverage, or a sample? If sampled, was it a representative sample?
  • For surveys, what was the response rate? Do the responders accurately represent the population? Are the survey questions reliable (consistently interpreted by different respondents) and valid (measure what they are intended to measure)?
• Are the data well organized and accessible? Is there a good user guide or codebook for the dataset? Can you import the data into a software package you are familiar with? If a dataset provides its own analysis tools, can you tell how they work?
• How were the statistics derived? Who derived/produced the statistics, and why? From what data source? By an appropriate method?
• Do the data/stats seem plausible? Does it fit with what you already know about the topic? Can you find independent sources that verify or are similar?