 Recently, there has been a lot of discussion over so-called sanctuary communities. These are places like California and Connecticut, Dallas County, Texas and Clayton County, Georgia, and cities like San Francisco, Seattle, Boston, and Philadelphia. In a sanctuary community, local resources are not used to enforce federal immigration law. Sanctuary cities have been called hotbeds of crime, but what's the real story? In early 2017, Thomas K. Wong of the University of California, San Diego, compared sanctuary counties to non-sanctuary counties. Wong found that sanctuary counties generally had lower crime rates, higher median incomes, lower poverty rates, and lower unemployment rates. In other words, communities that leave the enforcement of federal immigration law to the federal government do better than communities that try to enforce immigration law themselves. Wong's conclusion is based on a detailed statistical analysis of sanctuary counties, so designated by the Department of Immigration and Customs Enforcement. But many people are suspicious about statistics. Mark Twain famously opined that there are three types of lies. Lies, damn lies, and statistics. But the truth is, statistics don't lie. People do. And it's easy to do this, because if you omit certain pieces of information, a statistic is meaningless. Liars often omit this vital information. Including this information generally takes more than 140 characters. And what this means is that meaningful statistics can't be tweeted. Meaningful statistics can't be put onto an internet meme. And meaningful statistics can't be reduced to a 10-second sound bite. So let's check how Wong presented his data. So this is a table from Wong's report. And the sound bite statistic is, the violent crime rate per 10,000 is lower in sanctuary communities than it is in non-sanctuary communities. But if we look closer, we see that Wong also included some extra pieces of information. This N value and these P values. N is the sample size. It's the number of communities that were examined. So here there were 27 large central metropolitan communities that were sanctuary communities versus 35 that were non-sanctuary. But what's this P value? To understand what a P value is, consider the following situation. Suppose you're walking down the street and come across somebody who offers you the following game. There are two bags in front of you. One bag has an equal number of black and white tiles. The other bag has black and white tiles as well, but there is a substantial difference between the number of black tiles and the number of white tiles. You turn your back and the game sir gets rid of one of the bags. You reach in and without looking at them, you draw a handful of tiles from the bag. Based on this sample, you have to decide which bag you drew from. So for example, suppose you drew out 27 tiles and found that eight of them were black. That's less than one third of the tiles. And most of us would regard this observation as strong evidence that we drew from the bag where the number of white tiles was much larger than the number of black tiles. The P value is a way of quantifying the strength of the evidence and in this case, our observation corresponds to a very low P value about 0.05. As a general rule, statisticians say that evidence is statistically significant if the P value is 0.05 or less. Informally, statistically significant evidence should be considered very persuasive and should be considered strong evidence for a claim. On the other hand, suppose you drew 76 tiles and found that 44 of them were black. Here, while there are more black than white tiles, what we see isn't too different from what we'd expect to see if we drew from the bag with equal numbers of black and white tiles. So here, while there is some evidence that we drew from a bag with more black than white tiles, the evidence is rather weak. And this also corresponds to a higher P value, in this case about 0.17. Because of the weakness of the evidence for the claim that the bag had a higher number of black tiles, we might say that this observation is not statistically significant. Again, as a general rule, evidence that is not statistically significant should be regarded as not very convincing evidence for a particular claim. So let's take a look at Wong's data again. So for these large central metropolitan areas, the violent crime rate per 10,000 is 367.5 in sanctuary cities and 432.9 in non-sanctuary cities. Sanctuary communities seem to have lower violent crime rates. But if we look at the P value, we see that the P value is 0.038. It's a low P value. And so this is strong evidence for the claim that sanctuary communities have lower crime rate. And if we look down here for these micropolitan and non-core areas, we again see that the violent crime rate for sanctuary communities is substantially lower than the violent crime for non-sanctuary communities. And the low P values associated with these observations say that this observation is very convincing evidence for the claim that violent crime in sanctuary communities is lower. But wait, what about these large fringe metropolitan areas where the violent crime rate in sanctuary communities is higher than the violent crime rate in non-sanctuary communities? Well, if we look over at the P value, we see that there's a fairly high P value, 0.181. And remember, the higher the P value, the weaker the evidence. So here, there is some evidence that the crime rate in sanctuary cities is higher than it is in non-sanctuary communities. But the P value says that this evidence is weak evidence at best. If we examine the other data, we find that the evidence for lower crime rate, higher median income, lower poverty rate, and lower unemployment rate is statistically significant, which means that communities that leave the enforcement of immigration law to the federal government do better as a whole. If that's not enough support of sanctuary communities, here's another important factor. The US Constitution imposes strict limits on what states can and cannot do with respect to other states and nations. States can't print their own money, declare war, or tax imports. Only the federal government can enforce immigration laws. Only the federal government is paid to enforce immigration laws. Because local jurisdictions are not generally reimbursed for enforcing federal immigration law, when a city, state, or county helps the federal government, it essentially volunteers time, money, and personnel, taking these things away from other necessary tasks. Every hour a police officer spends tracking down an undocumented alien is an hour they're not spending solving a violent crime. Every hour a highway patrolman spends watching the border is an hour they're not spending responding to accidents. Every hour a social worker spends verifying someone's immigration status is an hour they're not spending helping a veteran. Detaining one undocumented worker for the federal government can cost local taxpayers more than $400 a day. Sanctuary communities believe that federal taxes already pay for enforcement of immigration law, and local taxes should pay for helping local residents. Being a sanctuary community isn't about enforcing immigration law. It's about getting what you've already paid for. Taxpayers shouldn't have to pay twice for the same service. Side effects of knowledge include greater understanding of the world around you and an increased demand of evidence from elected officials. Difficulty swallowing lies has been reported by some users, as well as reduced tolerance for flawed arguments, personal facts, anecdotal evidence, and false equivalences. Other users report becoming addicted to learning. A small percentage of users also exhibit behavior changes, such as fact-checking, rejection of internet memes, and political activism. Educational videos slow, but do not stop the spread of misinformation, propaganda, and alternative facts. If confronted with a serious case of these things, seek immediate help from nonprofit institutions of higher education or trained knowledge professionals, such as scientists, researchers, or investigative journalists.