Much as it did in the 2008 presidential primaries, the topic of undocumented immigration has become a dominant theme in the 2016 primary campaign, especially on the Republican side. Listening to the rhetoric from our candidates would lead us to believe that immigration patterns in the United States have remained largely unchanged over the past eight years. Recent analyses have demonstrated, though, that the immigration landscape that will face the presidential candidates of 2016 has changed considerably since 2008.
For example, it will come as a surprise to many Americans that after literally decades of growth, the number of undocumented immigrants in the U.S. has leveled off and even decreased slightly in the years immediately following the economic recession of 2007-2008. In fact, it is estimated that there are about one million fewer undocumented immigrants from Mexico in the United States now than there were in 2007.
Despite this new reality, a recent survey conducted by researchers (including myself) at Centre College in Danville, Kentucky has revealed that the vast majority of Americans believe that it is either definitely or probably true that "the number of undocumented immigrants in the United States has been growing significantly over the past few years." A full 50.1 percent believe that this is definitely true while another 30.4 percent believe that this is probably true. Only a small fraction of 13.1 percent (accurately) believe that this is probably or definitely false.
It is interesting to note that there is a partisan difference in responses to this question. Virtually all Republicans in our survey (96.2 percent) believe that it is definitely or probably true that the undocumented immigrant population has been growing over the past few years, while 74.2 percent of Democrats agree. This indicates that misinformation about immigration is clearly a bipartisan phenomenon, although the 20 percent difference between Republicans and Democrats may help explain why candidates like Donald Trump are making undocumented immigration a clear campaign issue, especially given that social science research has shown that perceptions of immigrant group size sometimes matters more than actual group size in driving immigration attitudes.
Another recent change in the nature of American immigration patterns is that there are now more immigrants in the United States from Asia than there are from Latin America, and this trend is projected to continue in the coming years and decades. This is especially noteworthy given that most Americans still associate "immigrant" with "Mexican," "Latin American," or "Spanish-speaking" which matters politically because attitudes toward Latin-American culture can affect attitudes toward immigration attitudes.
This same survey asked respondents whether they believed that "most immigrants coming to the United States today are from Spanish-speaking countries." 26.3 percent of Americans believe that this is "definitely true" and 33.5 percent believe that this is "probably true." Smaller proportions (accurately) believed that this is probably false (19.3 percent) or definitely false (12.7 percent).
Interestingly, survey results indicated that the partisan difference is not as pronounced on this question as it was with the immigration growth question, as 72.3 percent of Republicans and 60.7 percent of Democrats believe that it is either definitely or probably true that most new immigrants to the United States are from Spanish-speaking countries. (In this case the difference is within the margin of statistical sampling error.)
This is important because immigration policy preferences and priorities should (ideally) be based on an accurate understanding of the reality of immigration patterns in the United States. In this case, it seems that both the American public as well as our presidential candidates (with some notable exceptions) have some homework to do.
Conducted by Assistant Professor of Politics Benjamin Knoll and Associate Professor of Politics Chris Paskewich, the "Fall 2015 Colonel's Canvass Poll" was part of a community-based learning component of their fall 2015 courses at Centre College. In all, 77 students participated in fielding the survey and administering the questions to respondents. The randomized, nationally representative telephone survey was conducted September 24 - October 1, 2015. It sampled 487 respondents, 62 percent of whom were reached via landline and 38 percent via cellphone. The margin of sampling error for the survey is plus or minus 4.5 percent for the full sample and 7 percent for the partisan sub-samples.