By Leigh Thomas
PARIS (Reuters) - Marine Le Pen's protectionist, anti-establishment message resonated most in parts of France with lower incomes, lower life expectancy and lower education levels, a Reuters analysis of voting in Sunday's first round shows.
However there was no evidence that she scored particularly highly among young voters this time, many of whom backed hard-left campaigner Jean-Luc Melenchon. Moreover, it was not clear that she fared better in rural communities than in towns as some analysts have suggested.
Ahead of a May 7 runoff against centrist Emmanuel Macron, these are some conclusions that can be drawn by plotting Le Pen's scores in France's 'departments' - similar to counties - against government data on key social and economic factors.
Surveying Le Pen's results against more than a dozen such measures, the data suggest her message played particularly well in low-income areas with higher numbers of school dropouts.
In fact, for each 1,000-euro drop in the median income in a given area, Le Pen scored nearly two extra percentage points, according to a Reuters calculation using results and data from the Social Affairs Ministry. http://tmsnrt.rs/2pduHE9
Meanwhile, life expectancy for women tends to run nearly five years shorter than the average 85.2 in departments where Le Pen scored high and nearly three years lower than the average male expectancy of 79.1. http://tmsnrt.rs/2q1NmFS
But among the variables surveyed, education levels are the most revealing. The percentage of under-35 voters to have left school without a diploma has a heavy 60 percent correlation to Le Pen's scores. http://tmsnrt.rs/2q1PVb0
In statistics, correlation is measured between zero and 100 percent - the higher the count the closer the link.
One exception to that trend is Paris' northern working-class suburb of Seine-Saint-Denis, a traditional stronghold of the left, where over one-in-four 25-34 year-olds did not finish school. There, she won only 13.6 percent of the vote, coming in after Melenchon and Macron.
Unemployment is also a major factor explaining where Le Pen did best with the jobless rate by department 58 percent correlated with her scores. http://tmsnrt.rs/2pdpoo9
Departments with high levels of foreign-born populations tended to vote less for Le Pen, notably those areas concentrated around Paris and its suburbs. http://tmsnrt.rs/2p0iYdw
And although Le Pen scored high in some rural departments, that was not systematically the case. Similarly, although she did poorly in some more urbanised departments, she also did well in some more densely populated areas. http://tmsnrt.rs/2pdsEzY
Ahead of the vote, surveys suggested Le Pen was faring much better among voters under-25 than among the population at large, often by a margin as high as 7 percentage points. But in the end there was little evidence that age was a determining factor in her share of the vote. http://tmsnrt.rs/2q278Ry
A study by pollsters Ipsos of the first round results may explain why that was the case: while it found that 21 percent of 18 to 24-year-old voters backed her - largely the same as for the population as a whole - a massive 30 percent opted for her hard-left rival Melenchon.
(To view a graphic on 'French presidential election' click http://fingfx.thomsonreuters.com/gfx/rngs/FRANCE-ELECTION/010031D933E/index.html)
(This version of the story has been refiled to correct spelling of employment in bullet point and edits paragraph 11)