Estimating date / year of birth: Using "about" or "before"

Question

The WikiTree community has decided that an estimated date of birth is essentially required, particularly so that individuals can be shown and ordered within their approximate time periods. This is expressed in the guidelines, and arises when adding a new profile: "If you don't know the birth or death date for a person, you're encouraged to estimate a date... An estimated date is required when creating a new person if a reliable birth or death date is unknown." https://www.wikitree.com/wiki/Help:Estimated_Dates#Estimating_Dates

Rules of thumb are also offered for estimates. "For example, in the 1700s and 1800s in Northern Europe and the New England colonies, women often married at about 20-21 and men around 25. The first child was typically born within the first year of marriage." https://www.wikitree.com/wiki/Help:Estimated_Dates#Estimating_Dates

These vary somewhat for other time frames and regions, as described in Wikipedia and elsewhwere: https://en.m.wikipedia.org/wiki/Western_European_marriage_pattern

Following these guidelines, we apply whatever general as well as particular information we have to come up with a best estimate, and describe the basis for the estimate. But a question that often arises - and which is not directly addressed in the current guidelines but perhaps ought to be - is whether and when to use the indicator labelled "about" or the indicator labelled "before."

Some seem to use "about" even when "before" might be more likely to be and to remain accurate (i.e. unless and until a definitive date is found). For example, if what we know is that a child was born in 1600 - but we don't have info on the siblings - then we could apply the rule of thumb to suggest that the father was born "about" 1575 - or that he was born "before" 1575. If we know this to be the first child, then using about seems fine. But if, as is often the case with limited information, we do not know about the siblings, then using "before 1575" is much more likely to be correct.  

In particular, indicating that such a father's birth was "about 1575" could be off by several decades - since he could easily have been in his thirties, forties or even fifties when such a child was born. Using instead "before 1575" (or as I tend to do before 1580) is not only more likely to be accurate - since it was fairly uncommon for the father to be under 25 (and even less likely under 20 at the time) - but also the indication, even in the rare cases in which it is off, would only be off by a few years - not by decades.

In these situations, it seems both more accurate and more helpful to use "before" in the toggle field - and in the text to state "before about" (i.e. "before about X, based on Y").

Put another way, when we follow the rules of thumb based on a child of interest without knowing his or her siblings, then we are really making two assumptions when we use "about" rather than "before":

1. That the parent was at least 20/21 (mother) or 25 (father) when the child was born; AND
2. That this child was their eldest child.

This is particularly important for many of the immigrant profiles that I and others deal with for two reasons: 

1. In many cases, we know the immigrant (such as someone going from France to Nouvelle-France, or England to New England) and the parents named on their subsequent marriage record - but not their siblings (who in most cases did not emigrate); and 
2. In many if not most cases, the emigrating ancestor is not actually the eldest who might have title, land or another reason to stay - but rather a younger sibling who had none of those - and was therefore recruitable (as a soldier or "Fille du Roi" for example), or otherwise emigrated in search of an opportunity they did not have.

Finally, it should be noted that when the term "before" is used, the toggle field as it is currently set does not specify whether this is certain or uncertain. That is another reason to consider using "before about X, based on Y" in the text explanation. And when it is certain, the explanation can also address that, "before X" or "after X" based on a particular known event or record.

If anyone thinks this is not right please let me know - and in any case whether something about the selection of "about" versus "before" should be considered for addition to the guidelines. Thanks!

Comments

When you edit your post, can you please also add a note as to why it was edited.  Thanks.

---

Just a minor correction but I am planning to add a comment summarizing a few of the suggestions. Thanks!

---

I do a lot of GEDCOM uploads and it makes it impossible to compare people without a DOB or DOD. There are WT members that think if you don’t have a sourced date, leave the fields blank. I disagree with this. I never post a profile to WT without a date or country as I know a comparison is impossible without them.

---

I agree Dave. And as mentioned in a comment below, it's a lot easier to find and address the occasional duplicate than to have insufficient info or no profile at all - which tends to lead to someone "filling the vacuum" with potentially inaccurate info or mis-linking to another family or profile.

I've learned first-hand in historic projects that dis-entangling mixed-up families can be far harder than completing an occasional merge.

---

SUMMARY of Q&A and COMMENTS:

As promised, an overall summary of what turned out to be many helpful comments:

 

First, regarding the particular question and hypothetical raised (supplying a year of birth as opposed to specific birth/baptismal dates which are addressed below), many folks (including coordinators of projects I work with) seem to agree that it's fine to use before Year Y if that reflects what we know (and we then track that in the profile). The year picked with before can then be based on reasonable presumptions regarding the age of marriage and first children. For Nouvelle-France (Canada / Quebec to Acadia), the use of before Year Y based on a presumption that the mother was at least 20 and the father at least 25 is backed up by reasonably solid data (referenced below). And those years are not far from the WikiTree rules of thumb for Northern Europe and the New England colonies for the 1700s and 1800s.

 

Second, if we instead choose to go with about Year Y even when we don't know for sure where the child fits within the parents complete family (as reflected in the question) - then we could alternatively go with an estimated mean - but we should know about the actual probabilities of childbirths across the child-bearing period for the locality and time in order to estimate a particular mean or 'midpoint' year. And the resulting estimate using about would effectively be involving a second assumption, since the resulting midpoint year would effectively be to assume the child was in the middle, when we don't know that. Since the time frame is large and the estimate could off by quite a few years, it should be labeled as such, and potentially a flag used.

 

Third, it is also common to use before [specific date] DD-MM-YYYY for various purposes. Some users (but definitely not all) choose to use before in this way for noting a birth date based on a baptismal record. However, as has been pointed out, in many historic cases and places in which the baptism generally occurred very near to birth, it might well be the same day and not actually before. For these reasons, and when the dates are also generally so close if not the same, many people use the baptismal date (when that's all that is known), and simply note in the profile that the birth is based on the date of baptism.

 

Fourth, it has been pointed out that the WikiTree software for screening duplicates focuses on individuals within a couple of years of whatever year is referred to. However, that two-year screening window is also narrow with respect to our estimate if we use about - and since we would then be guesstimating the midpoint of a generally large range in our example, we might have even less chance of picking up a duplicate. Basically, this is another reason to encourage users to search for existing profiles first. There is now a very good WikiTree page to highlight that: [[https://www.wikitree.com/wiki/Space:How_to_Search_for_Existing_Profiles | How to Search for Existing Profiles]]

The good news is that when before Year Y is used, that appears very prominently for the people shown - and also at the very top of of their WikiTree profile (immediately following their name).

 

Finally, it has been noted (and the guidelines reflect), that we should avoid estimates based upon estimates. That is one reason to prefer before in the hypothetical (because it is based only on a presumption about the beginning of parents' typical child-bearing years and not a second assumption regarding where a particular child might have fallen in that range, be it early, middle or late). Second, when a particular event instead says it occurred about Year Y, it is not at all unlikely for the next person to suggest that if that was about Year Y, then a follow-on event must be about Year Y-minus-20, etc.. The use of before Year Y more clearly reflects that it's an open range yet to be defined. It guides research and thinking of that range and is not an indication that it should be the basis for further assumptions until better specified.

 

If anyone has any key summary points not covered, please don't hesitate to mention - and many thanks to all for great input and feedback!

Answer 1

I would only use "before" if I have firm date for an event.  For example, if I have a baptism date where I expect infant baptism to be the norm I would use "before" the baptism date as the birth date.  

In general I try to make an estimated date the best guess, or the middle of the most likely range because when another person makes a new profile for the same person it will only be flagged as a potential duplicate if the dates are within a couple of years.  Of course you can explain it however it makes the most sense in the biography.

Comments

I know some others who do the same Paige - in part because before is not marked as uncertain - but it is also not marked as certain. I think what you say about best guess is right - which is why I raise the issue. In a number of situations, saying before X will be more likely to be true than about X - and we can explain in the text what it's based on.

As an alternative, per your "middle-of-the-range" suggestion - if we should instead use about rather than before in my scenario - it would probably be better to suggest more 'loosely' that a parent was about 30/35 when sibs are unknown rather than that they were about 20/25. But as noted, each of these could actually be off by decades rather than years.

Answer 2

Even if about is selected, the biography should indicate that the birth date is estimated.

Comments

Indeed Mark - and I think even more so when it says "about year X" (rather than before a certain point based on parents' child-bearing likelihood)

Although both should clearly be explained

Answer 3

With respect to the quote from the guidelines:

"If you don't know the birth or death date for a person, you're encouraged to estimate a date"

I wish the last phrase said "you're encouraged to not add the profile until you can make at least an educated guess, instead of basing the person's age on the birth date of a parent or a child".  I also wish the guidelines clearly required the addition of the {{Date Guess}} template to the profile in such cases.

Comments

Date guess was replaced by {{Estimated Date}} -- which presumes what  is meant is birth.  Other parameters can be added, thusly:

{{Estimated Date|Death}}

{{Estimated Date|Marriage}}

{{Estimated Date|Birth and Death}}

{{Estimated Date|Birth and Marriage}}

{{Estimated Date|Death and Marriage}}

{{Estimated Date|Birth Death and Marriage}}

The Template's page also says : Always explain how you estimated the date in a Research Notes section.

---

I generally agree but point out that in many of these cases we're dealing with profiles that are in place because they're individuals of considerable interest, such as the families of pioneers who emigrated from Europe to the Americas.

I'm also reminded of the expression: "Nature loves a vacuum." So if we don't establish their profiles and provide the most relevant info available, they'll almost certainly be created by someone else, who may have less or incorrect info. Or, as in a number of cases I'm dealing with, the child instead gets attached into some other family with a similar name - which tends to become even more challenging than merging duplicates since various members of separate families eventually have to be manually 'dis-entangled' from each other. So in many cases, it seems much better to have the known parents profile linked to their child, along with whatever else is known about them in place.

In these regards though, we generally know who the parents are from the child's marriage or other record and place of origin. And we can make more than an educated guess regarding their required-to-be-entered birth, especially if we indicate that they're likely to have been born at least twenty years prior to their child. We or others can then add further information about them when and as available.

This also guides research - and potentially family connections in Europe - by providing their names, starting range of birth years, and locations (as well as anything else we might know about them such as other children, age of death, etc.).

---

To me the prhase "estimated" date means make an educated guess.  Making an educated guess does mean you use the clues you have, such as a child birth or a parents marriage.  In the eras I work, most parents were married, so a marriage year + one would be an educated guess if there was no other information to say it was later.  I would rather see a profile with an estimated date than wait for better information which may never be forthcoming or the creator of the profile moves on an never comes back to that person so they never get made.

---

When a parent's birth date is estimated as before 20 (or some other number) years before the child's known birth date, that doesn't sound unreasonable, but it gets very unreasonable very quickly when that parent's estimated birth date (note that in doing arithmetic, nobody looks at the "before" status indicator) is then used to estimate the birth dates of earlier ancestors.  After about 2 generations  of successive "estimates", you can easily be off by a whole generation, causing potential matches to be rejected because the dates aren't even close.

If we are going to encourage people to estimate dates then it should not be permitted to use already estimated dates as a basis for estimating other people's dates, otherwise it does much more harm than good.

---

We do have a style guideline that discourages estimating a date from an estimated date:

Help:Estimated_Dates#Avoid_making_estimates_based_on_estimates .

---

Lindy, I wonder how many people have ever consulted the style guide and, of those who did, I wonder how many remember seeing that, plus saying "discouraged" is not the same as saying "not permitted".  In fact, I'm exhibit "a" of that - I have read the style guide from start to finish several times and consulted it from time to time looking for a specific topic, yet until you mentioned this statement, I did not remember it being there.

I admit that my view is colored by having worked on an entire gedcom, where 5 or 6 generations had estimated birth dates.  Also, sources were not at all hard to find and I discovered that the guessed birth dates were off by as much as 130 years.

---

Thanks to Cindy for furthering the suggestion that having additional info is generally better than leaving info out!

And also to Gaille and Lindy for your thoughts. Reflecting on your experiences and comments, they strike me as an additional reason to consider the open range with before (when that's all that's known), rather than guesstimating with about. In particular, it's at least somewhat clearer to someone who sees "before 1650"  that since an open range is noted (rather than an approximate year) they should not assume the year is that, and use that to guesstimate further. 

On the other hand, when they instead see that one event is noted as occurring about 1650, then they might more naturally make a further jump, to say that the next event must be about 1630.

Clearly we can't prevent every faulty assumption - but it seems that the more we express what is really a range as an estimated year, then we likely induce more errors rather than fewer.

I also note that when the profiles are actually pulled up - which every user ends with - they clearly reflect that their birth is bef. X (indeed it appears right after the person's name at the top of the page).

---

My opinion is that the individual situation should dictate the choice of before/after versus about/uncertain. As long as the member provides a reasonable rationale for the choice, the member's choice should be accepted and respected. We should not remove the flexibility that our guidelines currently allow.

---

I agree.  Much depends on the quality of the estimate.  If it's based on relatively good information, for example an age at a census to determine birth year is a pretty good "about".  Also the entry about/uncertain does give latitude to be uncertain as to it's accuracy.  For the time I work, a 25 year old father would be relatively young, so a born "before" 20 or 25 years before wouldn't be out of line.  I would have a harder time using before 35 years even if that's the median age for parents, Tyler, as that would imply some accuracy that is not known.  If there is a child, it could be the first child or the 10th, and a born "before" 25 years ago still would not be wrong for the parent.

---

I can offer a real example of facing the "estimated date" problem on very large scale, but in this case it's death date, not birth date.  The Holocaust project has many profiles of people who died at the hands of the Nazis during World War II.  

There are lists of names of passengers on transport trains bound for a camp and we know the approximate date each train arrived at its destination.  A fair number of people died on the transport because they were literally packed in every cubic inch of space of cattle cars which were sealed for typically 3 or 4 days, during which they received no food or water and there were, of course, no bathroom facilities.

More than half of them were killed within a day or two of arrival at the camp and most of the rest died of malnutrition, forced labor, disease, or a sadistic guard at various times between their arrival and the end of the war, when the small number of them still alive were liberated by Allied forces.

So ... in almost all of these cases, they died between whatever date they were deported (the term used for sending them to a Nazi camp) and the end of World War II.  I asked a question here about how I should state death date and had several suggestions - after date of deportation, about date of arrival at camp (when that was known), about 1945 (when the camps were liberated), and before 1945.  There was no consensus and, unless I had a better guess, I mostly used before 1945 because it was the most certain of all other possible ways of estimating their death dates.  I explained it in the biography by saying "X probably died within a few days of arrival at Y camp, but definitely died sometime before the end of World War II."

---

I appreciate the input of all three of you pilots and very much agree. In fact I think we're approaching a rough consensus (here in and in some other comments) - even if using different starting points and examples.

Following up on Lindy's comment, I definitely agree that before or about can both be appropriately used. My added point (as reflected in the hypothetical) is that if we should feel compelled to go with about (I don't in cases such as the hypothetical), and we then see the guidelines as suggesting about 20/21 for a mother, we should note that it would be implicitly based on this also being her first child. If we know that, then the estimate seems fine. But if we don't know whether it's her first, middle or last child, then we would be making not one but a second big assumption by implying that she was about 20/21 when the child was born. It's for this reason that I think using before might be preferred in this kind of situation - since we're effectively then indicating (and should reiterate in the text) that while she was probably at least 20/21 when her first child was born, she could have been born earlier because we don't know, and shouldn't assume, that this known child was actually her first.

To Cindy's point, I not only fully agree - but it's the having to guesstimate with about (is the potential age more likely to be 20, 25, 30, 35, etc.) that makes me prefer to use before  if we don't know where this child fits, i.e. she was likely at least about 20 and he at least about 25 (if and when those numbers are appropriate as they are with our populations) - unless and until we know more. So unless we know more, I'd rather avoid the guesswork that about implies. And it has an added advantage besides likely accuracy being a range.  In particular, when we say "before x" - like before 1650 - a reader can't help but know it's the beginning of a range. As mentioned in another comment, if we instead say something like "about 1650" then first we need to emphasize that it's an estimate (and need to be sure we're considering the right midpoint in the example) - but it's also much more natural for a following person to conclude that: if that was "about 1650" then the next must be "about 1630" (all of this being reason to use the more supportable and informative before). 

Gaile - I can't help but appreciate the very "real" example you note. I think it's also a poignant reminder that with all due respect to ease of functionality for this or that, I think a higher goal is to be as truthful as we possibly can to the ancestors' lives that we're historically documenting. So if all we know is that they perished before a certain point, I'd much rather say that - and inspire further research - than guesstimate about something as important as their life.

Answer 4

In the scenario raised (which is not an unusual one), stating that the father was born about 1575 (i.e. that he was about 25 years old at the birth of this child, even if we don't know whether it was his first) - rather than suggesting that he was likely born before 1575 (i.e. that he was at least about 25) - highlights the issue.

In particular, the first (using about year X), might have a 20% probability of being true to within a year or two - and could be off by decades; whereas the second (using before X) might often have a greater than 90% likelihood of being true and would be off at most a few years - not thirty or forty years. 

I would call the first a guess (based on multiple assumptions) - and the latter just a basic presumption that they're at least of average child-bearing range (but we don't know more).

So if the system or instructions are somehow encouraging the former statement (with a combination of lower probability and greater range of error), then the guidance should perhaps be reconsidered.

Two possibilities for consideration are as noted:

1. Use before X (which is likely to be accurate and, if not, very close) - and clarify in the text what it is based on;
2. Use about X - but go to the middle of the range at issue (e.g. move the date from 25 years prior to birth to 35 years prior since we don't know the whole family) - and point out that it could still be off substantially, i.e. by more than a few years, potentially adding the {{Estimated Date}} flag as well.

In either case, I think it's helpful to make the statement as accurate as possible under the circumtances - and to detail the basis for the statement.

From experience with other sites and trees, I do think that providing the information with known names of parents (and any other info that's available), along with the basis for the year estimate, can help guide further research and avoid mistaken links - whereas omitting the known parents doesn't tend to lead anywhere, and sometimes has folks substituting 'other' parents. 

Comments

Many thanks to Melanie Paul for these added flags:

• {{Estimated Date}} -- which presumes what is meant is birth. 
• {{Estimated Date|Death}}
• {{Estimated Date|Marriage}}
• {{Estimated Date|Birth and Death}}
• {{Estimated Date|Birth and Marriage}}
• {{Estimated Date|Death and Marriage}}
• {{Estimated Date|Birth Death and Marriage}}

They seem especially relevant in the second scenario above: i.e. rather than using before with a "tight date" (such as father is presumed to be at least 20), going with a "guesstimate" (siblings unknown, father estimated to be 35).

And in either case as noted, the reminder:

Always explain how you estimated the date in a Research Notes section.

Answer 5

Sidestepping all the issues around estimating dates for events, an advantage of 'before' or 'after' is that it is easy to test a date using software. 'About' needs a range specifying as well.

Having said that, ‘before 1700’ may have a different implication than ‘before 23 March 1704'.

Comments

Date qualifiers like "before 1700" can be appropriate when there is a documented event that indicates that the person was of legal age as of a particular date, or (for date of death) when there is a date for the widowed spouse's next marriage.

---

I tend to agree with you Chris - and I think you're right in pointing out that when there is a particular date referred to (such as before 23 March 1704 in your example), that will generally be understood as something different than saying before 1700 - or before 1575 in my example (since this issue is especially problematic when trying to place some of our European ancestors with very little info available).

My principal concern in making any statement is that it should be as accurate as it can be given whatever the facts are (and the basis then explained). In that regard, using "before" allows for a reference statement that is most likely to be accurate and as noted even if not would only be off by a year or two.

It may be better to call this a "presumption" rather than a "guesstimate." So we can state that the father's birth prior to year X is based on the presumption that he was at least 20 for the baptism of his child - which in Northern Europe at the time covered most cases. If he turns out to have been 19 or less, then the year will still be very close and the statement still reasonably accurate.

I'm much less comfortable in using "about" for some of these circumstances. In the scenario raised, I would be guesstimating that the father was perhaps "about 35" when the child was born (siblings unknown) - when he could have been 20, 30, 40, 50 etc. So stating about 35 is then likely to not only be not true - but to potentially be off by a large amount.

---

I think our approaches agree!

When doing profiles for the Anglo-Scottish borders, and I have a child’s baptism source, as well as noting that it’s not a birth date in the bio, the ‘before’ 23 March 1704 implies birth is probably a few days before, though there are always exceptions with baptisms of older children. Then when estimating ages of parents, my assumptions are usually:

1. Both over 21 at marriage;

2. Man usually older than the woman;

3. Man often doesn’t marry until he has a land holding to farm

4. Average age of a generation is about 30, as per Aleš's statistics;

5. I round up ages to nearest 5 or 0 to avoid unjustifiable precision and allow easier spotting in lists

Consequently, mothers are 25 and fathers are 30 until more evidence turns up. (For those with a maths background, this is sort of Bayesian;-)

It works for me and that locale and period.

HTH, Chris

---

Chris - many thanks for discussion and spurring further research. There's a good article to suggest that for the pioneers that I and some others are dealing with, the mean ages for the parents at the birth of their middle child would be more like 30 for the mother, and 35 for the father:

New Estimates of Intergenerational Time Intervals for the Calculation of Age and Origins of Mutations https://www.sciencedirect.com/science/article/pii/S0002929707634375

 

But in fact, mothers had births from about age 16 to 48, and fathers from about 16 to 60.

 

This seems to highlight two key points. The first is that the estimates requiring a particular year when using "about" (rather than before) are both time-dependent and place-dependent. The second is that while determining the average generational span is clearly needed for DNA-based studies, and so the midpoint is important (as the paper reflects) - in the context of reflecting a single family, the assumption that a given child should be considered as a middle child is just as likely to be accurate or innacurate as the assumption that they're an eldest or a youngest. So the probability of error in using an "about" statement (which is very high given the extended range) is essentially the same when assuming it's a middle child (even if the scope of error might potentially be reduced to a decade or so rather than potentially two).

---

Chris, while it is easy to test a before or after date qualifier in software, the fact is that WikiTree doesn't do it.  None of the algorithms used to check for errors include consideration of those status fields.  These algorithms are so vague that, even with known accurate dates in the fields, we often see situations where a mother died the same year her child was born flagged with the explanation that a mother cannot die before her child is born.

---

I think that this is because dates are stored as yyyy-mm-dd but if the day or month are not known they are stored as yyyy-mm-00 or yyyy-00-00 which, in some cases, gives the peculiar results that you mentioned. E.g. if the day is zero and only the month and year is known,  ‘before’ should compare to the beginning of the month, and ‘after’ should compare to the end of the month (or the beginning of the following month).

I do not think that this happens in a consistent way across all comparison options.  I had thought of raising this, but thought Aleš had enough on his plate!

HTH, but I could be wrong, Chris

---

It is the WikiTree data validation process that does not handle "before" and "after" dates effectively. I don't believe that the same problem exists on the WikiTree+ features maintained by Ales.

Answer 6

When I added literally hundreds of Eckstädt-profiles, without knowing how they are connected to me, I once (but only once) had the case that I had to merge a couple of parent's profiles because I didn't realize at first that this is a parents couple with 2 documented children that are about 15 years apart. I use "about" unless I have a fix date when the person died or still lived. Then I use "before" and "after". And although a case like the one I mentioned above can happen, I think I will stick with "about".

Comments

Thanks Jelena - in many of these cases we're dealing with we actually know exactly how they're connected (and want to reflect that) - but we have very little info on individual pioneers from the 1600s - much less their siblings. Some of this is reflected in the question already but I'll add more.

Answer 7

I prefer to use "about" when I have a definite baptism or christening date.  Using "before" implies in the headings and search lists that the person was born before that year, which wouldn't necessarily be true.  I, of course, list the christening/baptism information in the bio. and sources. Just my opinion.  Anyone else?

Comments

Hi Faylene - I've also seen many profiles (on WikiTree and elsewhere) that use "about" when basing birth date on baptism date - when they're likely to be either the same or very close in time. 

I wish our "situation" was that easy though! In many cases with historic ancestors from the 1500s or 1600s, we don't have their baptismal record - if we did, that would be great and the end of story - following either your preference or that expressed above.

Instead, we often just have the birth of a key person (such as the family's ancestor emigrating from Europe to North America) and the names of their parents along with where they're from. So in the hypothetical (close to many cases we deal with in the 1600-1700s), we might know that a particular child was born in 1600. In many such cases, we know the parents from their baptismal and/or marriage records - but seldom their siblings (who would not normally be noted). 

So we don't know if this particular child of interest is their parents' first, last or somewhere in the middle.

If we simply presume that the parents are likely at least of typical child-bearing ages for the time, e.g. at least 20 or so for the mother (although usually the births were later), we could project her year of birth as being likely before 1580 - and we would cover the range of likely possibilities. So the statement is likely to remain true (even if this was not her eldest child).

On the other hand, if we instead pick "about 1580" / or "about 1575" / or "about 1570" etc. (since we don't know whether this is her eldest) - then whichever number we picked will likely be incorrect; and it could be off by many years, even decades.

Using before 1580 more closely reflects what we know is likely true - and directs further research from that point.

---

Thank you for your input Tyler. I understand what you have explained. If "before" is the preferred standard for baptism and christening dates for WikiTree and most experienced genealogists, then I will use that from now on.  It just bothered me that before infers no date of birth in the year they were christened or baptized. I guess most researchers will presume the "before" year is a probability for the actual year of birth.

---

Thanks for raising the point Faylene. And actually, in the projects I focus on (which tend to be more historic), the records tend to be sparse. But for the older Catholic ancestors in particular, baptisms tend to be quite close to the birth. In fact I've seen many in which the date of baptism is either the same day as birth or immediately thereafter.

So in many of these cases, your use of "about" seems perfectly right - and many do not reflect it as being before that day - since it may not be.

But as mentioned I wish more of our historic situations were like that. We're often dealing with much less available info - and in many of those cases, using before year X to reflect the likely range seems better and more informative than stating it's about year X (without sufficient support).

---

Your point well taken Tyler.  A birth date surely wouldn't be "after" the baptism or christening date. II too have seen many christening and baptism dates being the same as the birth date or shortly after. "Before" it will be. 

---

Both actually seem okay in that situation Faylene, but as Jean notes below and I tend to agree, many folks and projects just use the date reflected in the record cited (which in many cases is a baptism that could be the same day or immediately following). 

We can then state what's known in the bio: "[Person x] was baptized on date X - or born on date X" - depending on which we know.

---

Tyler, Yes. I always use born, baptized or christened on x date in the bio, when known. I agree with Jean's notes below and your reply to her.  Sounds good to me. Thank you!

Answer 8

hmm, there was another discussion recently about dates  https://www.wikitree.com/g2g/1327747/how-to-mark-profile-needs-dates

The ''before''  and ''after'' toggles don't actually do much in my experience, and do not transfer over if the data is picked up elsewhere.  Which happens; before coming to WikiTree I was using MyHeritage which would offer WikiTree data ''for free'', one could then add it to our own tree.  When I switched to WikiTree instead, my GEDCom import totally lost those ''before'' and ''after'' indicators.  If they were even present when I took the data in the first place.  Too long ago to remember.

Also, when merges are done, those toggles get lost in the shuffle.  So you get just the date with no indicator about around, before, after.

Comments

The before, about, and after qualifiers do show up on the merge screen when I am completing a merge. If they didn't carry through to a merged profile, it is probably because the person who did the merge chose not to carry them over.

---

Ellen - I'm glad the date qualifiers generally appear!

But Danielle - you make a good point about them sometimes not being carried over into WikiTree - which no doubt explains why they're sometimes absent from the profiles we're working on.

The question then is what do you tend to do in the scenario raised since you've certainly seen, and worked to improve, many early ancestor profiles!

---

Just took a look at the BALSAC study you linked, it is post-conquest in its time period.  So age of majority was 21.  Would be interesting to see a similar study from the data collected for Cartagène study, which is much more extensive.  (I took part in that, included full genome sequencing for about 30k people, plus family trees, we don't get individual results, unfortunately).

But for the specific case of migrants from France to New France pre-conquest, realize that the age of majority was 25 then.  Legal cannonical age at which girls could marry was 12, for boys 14.  Very few boys marry that young, they are not established and therefore not in a position to support a family.  This age range does not necessarily apply to other European cultures.

So for migrants in the early days, I will estimate the father's age as at least 25 years before the migrant's dob (per whatever data we have, such as census records etc), and the mother's age as 5 years less than the father.  I always note that this is an estimate on the bio, although I don't use the banner, find that extremely ugly, wish they would change it to something else.

---

Thanks Danielle - suggesting that the father's age was likely at least 25 years and the mother's at least 20 (or reflecting before 1575 and before 1580 in the hypothetical) seems to make the most sense overall.

And I generally dislike the added banner as well since it dominates the entire profile and makes it look "defective." But I think if we're just applying a basic presumption that the parents are at least of typical child-bearing years (or older), then the "guesstimate" banner is not as pertinent. I think that saying "before X" is already helpful in that regard because it alerts the reader that it's open and there's a range.

Also, what's the Cartagène study and is anything published yet..?

---

https://www.cartagene.qc.ca/en/home

genetics study, mostly medical research.  And yes, there are various things published, have gotten some of them sent me (via e-mail link).  They switched to Covid research lately, for obvious reason.

---

Thanks - I'll take a look at Cartagène as well!

Also the study published from BALSAC started from a group of later families but then extended them backward in time to founders as well - as part what they called the "Genealogical Sample": 

These genealogies have an average depth of 9 generations, but many lineages go as far back as 12 or 13 generations. Overall, 237,822 ancestors were identified in the 100 genealogies, corresponding to 15,752 distinct individuals.

They also did a separate analysis tracking back just the genealogies including marrying "founders" (their cut off was basically folks who married in the 1600s):

Among ancestors with a recorded date of marriage, we identified 457 distinct founders who married before the year 1700. Intergenerational intervals in the descending lines of each of these ancestors were measured. The distribution of these ancestors according to the mean interval length of their descending lines is presented in Figure 5. Again, most of the mean interval lengths hovered around 30 years: 68% were between 28.5 and 31.5 years. None of the ancestors’ descending lines had a mean interval length <25 years, and some of them reached as high as 37 years.

There was an observed shift (to a couple of years earlier) - which was at least partly attributed to the fact that the most frequently appearing ancestors were founders - and with a shortage of women in the earliest decades of the colony there was a somewhat younger age of marriage. I suppose the early recruitment age in the "Filles du Roi" program could well have contributed also.

The presence of such variables affecting what would be the estimated mid-point of the appropriate range or "mean interval" in their terminology (not only for these populations but more broadly) seems to underscore the benefit of going with a basic and easy to use presumption. In this case, as suggested, that the mother and father were generally at least 20 and 25 respectively for the groups that we're dealing with. 

The data in the paper are actually fairly supportive of that general baseline for a range, even more so for the founders. For example, the mothers' mean age at first birth overall was just under 23 - but it was somewhat younger among founders - so I'd say our working base presumption that they were generally over 20 seems pretty close to spot on.

---

from Yves Landry's book on the filles du roi, one thing he mentions is that age at marriage of girls in France was later than for girls here in the same era, for the obvious reason that there were so few girls available here.  So girls getting married here at 12-13 in early days is not unusual, but as the population grew and the sexes equalized, this got pushed back to older girls, although still lots married in their teens.

---

Girls getting married at 12-13 would certainly skew the average Danielle.

In fact, it's hard to really imagine what some of our ancestors went through - but in any case we owe our presence to them!

Answer 9

if I have a baptism and not a birth I do not use either as many baptism and births can be on the same date. In those cases before and after is then not correct. That it is the baptism date is clear in the profile.

Comments

I tend to do the same when we know the birth is likely very close and we don't actually know that it's before the date of baptism. It's not unreasonable to use that date and it seems that many of our profiles and outside projects do just that. 

And as you note, I just start the bio with "[Person x] was baptized on _" / or "...born on" (if I know that) - which makes it very clear.