It's remarkable how you can know some much about our elections from such a distance. Isn't the internet wonderful?
OTOH as I said the increasing latino vote that voted 66% for Obama, made a marginal difference in several important swing states. Of course many of the Latinos are also "young". 8)
http://www.reuters.com/article/2012/11/08/us-usa-campaign-diversity-idUSBRE8A62MO20121108
Of course in a race this close (Obama only got 53%) almost any group could get credit or blame.
http://www.nytimes.com/interactive/2012/11/07/us/politics/obamas-diverse-base-of-support.html
Per the NYT
"Young voters favored President Obama,
but less so than in 2008. However, he managed to improve his share of the youth vote in swing states like Ohio, Florida and Virginia where his campaign most actively targeted voters."
"President Obama won the Hispanic vote by 44 percentage points, 8 percentage points more than in 2008. Among the swing states, the president made the biggest gains in Colorado, taking 74 percent of the Hispanic vote, up from 61 percent in 2008. In Florida, President Obama’s gains among Hispanic voters helped him in the state. He won 60 percent of the Hispanic vote, up from 57 percent in 2008 and 44 percent for John Kerry in 2004."
Of course the same data can be parsed different ways. Young/old, Black/white/other. Just like in macroeconomic studies the experiment has multiple variables that can not be isolated and completely controlled for.
JR