Should we DNA Test the "Older Generation"?

I am frequently asked, "What DNA test should I take?" Recently my answer has been, "What genealogical question are you trying to answer?" As often as not the inquirer has not thought about that question. However, if one tests and then tries to figure out what to do with the results, frustration is likely to result. The test chosen may not be appropriate and the right family member(s) may not have been tested.

Within our bodies we have a rich source of information about our family history. Sometimes it can provide answers when oral traditions and written documents fail us. However, DNA testing must be used intelligently in order to tap into this abundant source of information. DNA can help confirm or refute what we think we know about our pedigree chart. It also has the potential to connect us with relatives that were previously unknown to us.

As most of you know there are three types of commercially available DNA tests designed for furthering family history research. Two of the three, Y-chromosome and mitochondrial, are available at different levels and different costs. Only males have Y-chromosomes. Everyone has a mitochondria but men do not share theirs with their children. These two tests have the power to give us information about our ancestors going back a few hundred years. However, each only gives us information about one direct line on our pedigree chart. The Y-chromosome test gives us information about the top (paternal) person in each generation of our chart. The mitochondrial test gives us information about the bottom (maternal) person in each generation of our chart. Even if one's chart only goes back four generations, there are beginning to be a lot of ancestors about whom we are getting no information.

The third type of DNA being tested for genealogical research, autosomal DNA, has the potential to give us information about all the lines of our pedigree chart. The trade off is that it cannot provide this information on everybody to whom we are are related for more than a few generations back in time. Below is a chart from page 163 of my new book Crash Course in Genealogy:

As I am sure you noticed, the probability of finding all your potential cousins drops off considerably beyond the third cousin level. The above probabilities are a combination of the experiences of FTDNA and 23and Me in their first few months of testing in 2010. One experienced researcher claims more recent experience shows the probability of finding second cousins is higher than 99%. My main point is that when it comes to autosomal testing the generation of the person being tested does matter. For example, if your grandfather could be tested, he should have a much higher probability of matching with your fifth cousins who are his third cousins (or third cousins-twice removed) than you  would.

I have now decided that there are two times one should order a DNA test kit. The first remains when you have a genealogical question that it might help answer. The second is when you can capture the DNA pattern of someone in the older generation while they are still with us.