Immigration Myths and Statistical Literacy

Over at CNN I noticed an opinion piece by Fariborz Ghadar titled “Dispel the immigration myths.” This is an interesting piece in that the approach taken by Ghadar is data driven. Thus one could argue that by addressing the statistical literacy of the reader on the immigration issue he is dispelling some of the myths surrounding the debate in immigration policy. Unfortunately the use of data in the piece badly fails the test of statistical literary. He mixes percentages and level in inappropriate way, used data that is not relevant to the issues at hand, and he makes comparisons that are not valid across countries.

I am not trying here to take a position on immigration or on Ghadar’s options on immigration issues, but rather with the methods he uses to make his points. I will also not exhaustively examine every statement he makes, but rather only deal with some of those he makes early in the piece. The point is if someone is going to make a case for any argument they should use the appropriate data and use is wisely.

The first “myth” that Ghadar tries to deal with is that “America continues to be a nation of immigrants.” Referring to this link he points out that we are currently at about 13 percent immigrants in the United States. By that he means that 13% of those currently living in the country as of 2011 were not US citizens at birth. The graphic, as it was posed the day I looked at it had a problem with the labeling of the axis so I have reproduced the relevant part of the graph at the right. the data covers the period 1850 to 2010. The more recent 2011 numbers puts the percent immigrant population at 13.0 percent. All of the data are based on US Census Bureau published figures.

Ghadar’s statement is “While America at one point had a huge influx of immigrants, today the picture is quite different.” But using his numbers is it really all that different? From 1880 to 1920 the percentage was higher, but it peaked at 14.8 percent. If we were a nation of immigrants at the 14.8 percent level is 13.0 percent all that different. At those levels about one in six or seven people you meet would on average be an immigrant. That is not to say that being a nation of immigrants is a bad thing. It is just that the current situation is not that different than that of the the early 20th century. The more relevant questions would be to ask if 13 percent of the population being immigrants a problem?

The second “myth” Ghadar address it “that everyone who is thinking of emigrating wants to come to the U.S” Given the migration of populations around the globe it is hard for me to believe that anyone thinks they all want to come to the United States. So I do not see this as a “myth” that needs to be dealt with.

What does Ghadar say? He points out “As a percentage of population, Canada and Australia both have significantly higher rates of foreign-born residents than the U.S., at approximately 20% and 26% respectively.” He goes on to say that the rate (13.0) “is now on par with those of France and Germany.”

A problem here is that Ghadar is mixing percentages and numbers. It is not a question of who has the larger percentage immigrant population but rather of what proportion of the immigrants are going to each country. So how do the countries compare. The United States with a 2011 population of 311.8 million would have approximately 40.5 million immigrants. Canada with a 2011 population of 33.5 million would have approximately 6.7 million immigrants, and Australia with a 2011 population of 21.8 million would have an immigrant population of 5.7 million. Thus in terms of the number of immigrants the United States far surpasses both Canada and Australia. Ghadar made the error of looking only at the percentages and not at the number of immigrants as he should have.

But there is a second more fundamental problem with Ghadar’s approach. The current level of immigrant population is not a measure of current preferences for countries by the current immigrant population. Those are the people Ghadar is talking about. The percent immigrant population is a measure of immigration over the life time of the current residents of a county. That includes those who move to the county decades ago. He needs to find a very different data set to make his case on this “myth.” The relevant data would be on immigration in the last ten year.

Enough said on the specifics of the opinion piece. If one is to make a case for a specific position then they need to use the right data and use it correctly. This is the essence of statistical literacy. Unfortunately that was not done in this piece.