This particular article discusses the predicted voting habits of Latino voters in this year's presidential election. I found a few issues with the wording of the article or the reliability of it, especially in the behavior vs. response department and the fact that data seemed to be manipulated to appear more optimistic.
A quote from the article that I found misleading: "A survey released this week indicates 87 percent of Latinos nationwide are "almost certain" they will vote next Tuesday."
I found this misleading because the title of the article is "Survey: 87 percent of Latino voters will go to polls." This was a manipulation by the author to make readers more shocked and intrigued. Since the actual response was "almost certain", that really does not tell us much. I could tell you that I am almost certain I am going to eat at a restaurant tonight, but there may only be a 30% chance in reality. This response is a bit in the grey area, as many people will have a different perception of what almost certain actually means or how sure they should be do mark this answer. Also, if 87% are almost certain they will vote, their behavior can prove to be very different on election day. Something else may come up, they may forget, or people may not be able to still register due to time constraints. There are a lot of excuses that could prevent these 87% of those Latinos polled to not go.
Also, another problem is that there are a lot of illegal immigrants from Latin American countries here, which, most likely, were not included in this survey as they can not vote. Therefore, the actual percentage may be skewed since a significant portion of the total Latin American population is actually left out of the data from the start.
Another issue I found was in the statement, ""The poll shows that this year we can anticipate record participation among Latino voters," impreMedia CEO Monica Lozano said in a statement." This could be misleading data in that this year our country has more Latinos than in other election years. This is by no means a bad thing, but it is simply misleading with this data, unless they are discussing percentages of Latino voters, in which case the speaker should really specify.
This survey seems to have content validity in that it measures what it should measure: the outlook for Latino voters, but how reliable and re-testable is this data? It seems hard to say, as often political surveys change the most from day-to-day and it is hard to actually predict voter behavior prior to the election.