Studies for Political Purposes in regards to Crime, Poverty, Economics, and Healthcare can be highly inaccurate as the differences between a Big City and its Suburb can be significant.
In fact, larger area studies, like State-by-State, are also highly inaccurate as a single city can easily skew the entire state's numbers as the geographical area is way too big.
Additionally, even the suburbs can easily make bad inner cities look good.
Statistics can be manipulate via changing the time range of the data involved..
Let's use Housing Prices and the Economy as an example.
Since Housing Prices & the Economy can go up & down, a research study can pick a year in the past that starts off low (or high) to make future gains (or losses) look good (or bad) with respect to a public policy, Social Carrot program or elected official's effectiveness, e.g. Pick a starting year before (or during) a housing crash (or boom) and pick a ending year that fits your polltical goal.
STATE BY STATE, CITY by CITY, COUNTER, by COUNTY:
INCOME TAX COMPARISONS = MEANINGLESS STATISTIC
Because 50% of all tax payers don't even pay income tax, this statistic is completely meaningless. Any income tax comparisons between states, cities and even counties, are only comparing the income tax of the rich as those are the one's who actually pay income tax.
