I completed a DNA test, and the results stated that I am 77% African 20% European 2% Native American and 1% Asian.

From the results of this test, is there anything in the results that would tell me what side of my family (such as one of my grandparents or parents) I inherited my 20% European ethnicity from?

  • 2
    Welcome to G&FH.SE, I edited your question to be more general and ones that would conform to the community better but still would facilitate the root of your question. Could you please edit your question to include where you had your test done and what test you had done as each service interprets the ethnicity slightly differently? When you have a chance please also take the Tour to understand our format and our rules.
    – CRSouser
    Jun 11, 2015 at 14:24
  • 2
    I see you posted this question again (technically for the 4th time) so perhaps you didn't see our responses. All of the answers below address your originally asked question which as exactly asked does not conform to community standards. If you are looking for a short answer, the answer is NO and it is best you do additional research beyond DNA to determine this as at 20% it should be relatively recent. I would also suggest you upload your results to GEDMATCH for further 'matching' as well as for $10 you get access to their extended ethnicity tools if that is your primary question.
    – CRSouser
    Jun 12, 2015 at 16:17

3 Answers 3


There's an underlying problem with this question (both as it was originally posed, and in its edited form), namely, it assumes that the ethnicity estimates can be applied to the paper-trail genealogy in this way.

First one needs to consider if the ethnicity estimates are accurate, or even if they can be. Companies use these results to engage new customers, but is this information useful, and if so, how useful is it?

Before starting off on this kind of analysis, it is important to consider which company you've tested with, what test you've taken, and what that company's track record is. The user must weigh the marketing claims against what it is actually possible for the science to tell us, otherwise you're just letting yourself be led down a false research path, in the same way as accepting a same-name Ancestry shaky leaf for someone who is not actually your ancestor.

Judy G. Russell sums up the situation in About that sharing button…, posted on 4 October 2015 on her blog The Legal Genealogist.

Let me repeat, once more, that we have to keep in mind what these admixture tests do: they take the DNA of living people — us, the test takers — and they compare it to the DNA of other living people — people whose parents and grandparents and, sometimes, even great grandparents all come from one geographic area. Then they try to extrapolate backwards into time. Nobody is out there running around, digging up 500- or 1,000-year-old bones, extracting DNA for us to compare our own DNA to.

Ancestry's marketing department is (in my opinion) particularly bad about overselling their features. In this post Russell points out that Ancestry is offering customers canned messages when they share their results, which don't say anywhere that these are estimates, and estimates with serious limitations.

Below are more blog posts where problems with ethnicity estimates and admixture results have been discussed. I will post these in reverse chronological order (newest on top). Here's why: the companies refine their tests, so what they report as the results to their customer base is going to change over time. If someone reports that Company A's reporting is seriously broken in 2012, that doesn't mean it's going to be that broken forever. But if a user had asked your question in 2012, that's what the test would have shown them. So you have to ask: how much weight do you want to place on the assumption you've just made as you carry on with your research, knowing that a couple of years down the line, those percentages may say something else?

It is important to remember that as intriguing as these admixture predictions are, none of them are 100% accurate at the granular level. We still have a long way to go before anyone can honestly claim to be able to tell a person exactly where their ancestors once lived based on their autosomal DNA alone. However, we are making progress and in a couple of years we will likely be amazed at the advances. In the meantime, the ever-increasing competition between the major companies is proving to be beneficial, spurring all to improve their offerings. So, don't count any of them out quite yet. In a short time, my opinions regarding the "best of" may have completely changed again!

These links may not directly address the question you asked, but they do show why it is important to know which company's ethnicity estimates or admixture results you are looking at and which test they came from, in order to give you a better answer to your question.

Category Links and general resources:

For those already familiar with DNA: Blaine Bettinger's attempts to re-create his grandmother's genome with a Tier 1 tool at GEDmatch called Lazarus are detailed here:

Presumably -- if you had enough relatives that were willing to test -- you could answer the question of what parents or grandparents your European ancestry came from by using this kind of analysis.

Roberta Estes explains how it works here: Lazarus – Putting Humpty Dumpty Back Together Again (posted on 14 January 2015)

  • 3
    Keep in mind that genetic "admixture" is a popular term, but is often used inappropriately. It is not simply two people of different ethnicities interbreeding - it is a population concept and is therefore only appropriate when used in reference to many (i.e. populations of) DNA samples. Since this answer focusses on admixture, just to be clear, admixture does not occur simply when a European person has offspring with an African person.
    – Harry V.
    Jun 13, 2015 at 8:30

Although 20% European is a significant proportion, I doubt the answer is as easy as all of it coming to you from single grandparent.

More likely, more than 1 grandparent (each contributing about 25% to you) was already mixed ethnicity. This might verifiable in other records. It is interesting that the 3% Native American + Asian component is 1/32. It might all come from a single 3xgreat-grandparent, or also be from multiple ancestors further back.

Keep in mind, the ethnicity results are estimates, based on current algorithms using current reference databases - subject to inaccuracy and (hopefully) ongoing improvement. Also autosomal DNA testing is not gender-specific; there is no way to determine which ancestor contributed which bits without both matching to other testers AND a record trail.

There are a number of resources cited in the answers to other DNA-tagged questions here; please check them out, especially Generational Loss of Data with DNA Testing


The DNA tests are usually quite accurate, and it is likely that one of your genetic grandparents indeed has the differing ancestry.

If you are in the situation that you believe you know who all your grandparents are, and that the ethnic connection does not seem possible, then this is likely what is best known as (in genealogical DNA analysis) as an NPE event, i.e. Not the Parent Expected.

There are several different ways this can happen: adoption, infidelity, or a crime of a sexual nature. Often, especially for our grandparents' generation, these events were covered up and children were raised as if they were the true children of both parents. In some cases, the other parent may not even know they were not the true parent.

Yes you can find out which grandparent was not the true grandparent. Before you do so, ensure that there is nobody who would be hurt by this knowledge, and that your family approves that you investigate. Once you open up a can of worms, it can't be closed again.

If you are satisfied that finding the answer will not hurt or offend anybody, this is what you can do:

First, check your own results:

Find out through the DNA testing company you worked with, what other testers match some of your DNA and seem to be related to you. If you know they are related, or if you can see by their research that they connect to one of your grandparents, then you can be fairly certain that their connection is through that grandparent.

If you have enough matches from that DNA service, you will likely find that 3 of your grandparents have matches and one doesn't. The other people are likely related to your true genetic grandparent, and they'll be people you don't recognize as relatives. That "blood" grandparent will likely be the one who is supplying the different ethnicity. The grandparent without matches was likely the one who raised your parent, and may or may not have known that they were not the genetic parent.

Second (if necessary), go the extra step:

If you don't have enough relative matches from your own DNA test to determine the grandparent, and you still are confident enough that disturbing the family history will not cause animosity, then you can do this: Find one of your cousins that you know/think is related to each of your 4 grandparents on only that grandparent's side who is willing to do a DNA test.

When you get the test results, the relative who does not have a 1/4 match to your DNA results will indicate which grandparent is not your genetic grandparent. Finding common relatives between that cousin and yourself will help you start to build a family tree and ultimately the identification of the unknown grandparent.


You can do this before or after. And you don't have to tell anyone either: Find pictures of your 4 grandparents. Find pictures of their children or grandchildren. See if you can identify the one grandparent that has the fewest physical features in common with the children and grandchildren. If one grandparent doesn't seem to have passed on their appearance, they likely may not have been the genetic grandparent.

  • Downvoted because DNA ethnicity admixtures are not that accurate, so the answer is misleading.
    – user104
    Jun 13, 2015 at 19:36
  • 1
    @ColeValleyGirl - They're accurate enough to say that 20% European is significant and indicative that a grandparent or great-grandparent may be an NPE when no European descent was expected.
    – lkessler
    Jun 13, 2015 at 23:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.