The study, published on Wednesday, found that with each new child parents were increasing their likelihood of falling into poverty.
While 13 percent of couples with one child were threatened by poverty in 2015, this rose to 16 percent in two-child families and to 18 percent for couples with three children.
The study took households which earn less than 60 percent of the national average income to be at risk of falling into poverty.
Previous studies by the Organisation for Economic Co-operation and Development (OECD) into poverty among families had come to much less startling conclusions about the seriousness of the problem.
According to the Bertelsmann study the OECD methodology distorts reality, making poorer families appear to be richer and richer families appear to be poorer.
To name one example, the Bertelsmann study found that 68 percent of single parents are at risk of poverty, a much higher proportion than the 46 percent the OECD reckons fall into this category.
The Bertelsmann study also concluded that the gap between richest and poorest households has widened more significantly since 1991 than OECD research suggests.
The decisive issue in the data analysis is the so-called "equivalence scale." The scale is used in order to compare households of different sizes. For example, a family of four people has the advantage over someone who lives alone that they do not need four dinner tables to achieve the same state of comfort. The equivalence scale aims to take such savings into consideration.
The OECD scale gives a household a weighting of 0.5 for every extra member over the age of 14 and a weighting of 0.3 for extra members under 14. But the Bertelsmann researchers concluded that these weightings were too arbitrary and came up with a different scale after conducting closer examinations of the disposal incomes of the households at various points on the income scale.
Not everyone is happy though, that the Bertelsmann Foundation have proposed throwing out the OECD scale.
"The discussion about the methodology has been going on since the 1990s," says OECD expert Michael Förster, arguing that starting methodological discussions again would be a step backwards. He said that, while the Bertelsmann study's finding are broadly correct, "we only come to different results in detail".