oftentimes studies based on collected statistics fail to recognize an important factor of the study. it may be either too difficult to account for, or simply forgotten. statistics that have any validity must include a large base of data, but there are many other issues involved with analyzing accurate trends in data.

how often do you hear that men are worse drivers than women? they are involved in more accidents, the statistics show it... insurance rates show it. the data that insurance companies use is from a very long timespan. it also includes a time when women drove less often than men, simply because men drove more often (that's another node). a more accurate study would analyze the number of collisions per mile driven for each gender. this is an interesting example of the abuse of statistics.