How reliable are dating methods?

[Micaiah]

If every dating method gives you a date of, say, 2.2 billion years, then the overwhelming probability is that 2.2 billion years is its date. When you have the evidence clearly observed, then and only then can you apply a theory to explain why the evidence is saying that it's 2.2 billion years old. And that theory must fit all the evidence, including explaining why the dating methods used can't be wrong. If it doesn't explain all the evidence, then it is an invalid theory, not just an alternative reading.

How reliable are these dating methods ?

 

[notto]

Very. When used properly and within the limits of each method, they provide results that when cross checked with other methods and samples of known age show the consistency of the results.


http://www.asa3.org/ASA/resources/Wiens.html
http://www.soest.hawaii.edu/GG/HCV/haw_formation.html

 

[Micaiah]

Say I go out and test some rocks and find they have age of 1000 billions years. What degree of accuracy would you say is required to be able to say the dating method was reliable.

 

[rmwilliamsll]

Say I go out and test some rocks and find they have age of 1000 billions years. What degree of accuracy would you say is required to be able to say the dating method was reliable.

since there have never been any rocks on earth dated much older than 4* billion years i would suspect that an answer of 1000 billion is really wrong. as in 250 times way too high. but i suspect you pulled the number out of the air and are really only interested in the error or reliability band for dating. that answer appears to be +-.1 billion. see: http://www.gate.net/~rwms/AgeEarth.html

*i believe the oldest currently are at 3.7-3.8
http://wrgis.wr.usgs.gov/docs/parks/gtime/ageofearth.html
but it is not my field so my info might be outofdate a little
.....

 

[notto]

Say I go out and test some rocks and find they have age of 1000 billions years. What degree of accuracy would you say is required to be able to say the dating method was reliable.

What do you mean by accuracy? I think you are talking about consistency. Independent testing methods are consistent with each other and are tested against each other. The consistency of the dating methods (which use different physical processes and properties) shows that they are accurate because they are consistent. The methods are empirical and repeatable.

I don't think I understand your question. The reliability of the method is determined by how its results are consistent with other methods or consistent with empirical tests against samples of a known age.

Accuracy is given in measurements in the form of an error measurement. The accuracy depends on the statistical and mathematical calculations and is usually given as a +- as with other emerical statistical measurements.

 

[Crusadar]

Micaiah said: Say I go out and test some rocks and find they have age of 1000 billions years. What degree of accuracy would you say is required to be able to say the dating method was reliable.

Actually there are other factors that must also take be taken into consideration. You can't just date any old rock, you need to also know where it was found (i.e. the strata) and the type of fossils that was found with the rock - otherwise you will get dates that are all over the place and wouldn't know which date to pick as the correct one. What always bothered me was that when rocks with know ages are dated, dating methods don't work, while rocks with unknown ages are dated, dating methods do work! Now you tell me if you can accurately date anything based on that assumption!

 

[notto]

Micaiah said: Say I go out and test some rocks and find they have age of 1000 billions years. What degree of accuracy would you say is required to be able to say the dating method was reliable.

Actually there are other factors that must also take be taken into consideration. You can't just date any old rock, you need to also know where it was found (i.e. the strata) and the type of fossils that was found with the rock - otherwise you will get dates that are all over the place and wouldn't know which date to pick as the correct one. What always bothered me was that when rocks with know ages are dated, dating methods don't work, while rocks with unknown ages are dated, dating methods do work! Now you tell me if you can accurately date anything based on that assumption!

Unsupported claims. Can you provide some details on this? You claim that dating methods do not work when rocks of known ages are dated. This is certainly true when creationist use poor methods of collections such as dating xenoliths within the sample but other than that I think your conclusions are wrong.

Fossils are not needed to date rocks. We don't find fossils in volcanic rock. Your just putting forth creationist falsehoods now.

Can you show us an example of this 'all over the place' dates you are referring to? Please be specific and include reference to the methods used to do the dating, how the samples were collected, and how contamination was kept out of the dating.

 

[Micaiah]

My mistake. Thanks for your kind correction. I meant 1 000 000 000 years, so that would give an accuracy of +- 100 000 000 years. That is pretty accurate in my view.

 

[Micaiah]

Micaiah said: Say I go out and test some rocks and find they have age of 1000 billions years. What degree of accuracy would you say is required to be able to say the dating method was reliable.

Actually there are other factors that must also take be taken into consideration. You can't just date any old rock, you need to also know where it was found (i.e. the strata) and the type of fossils that was found with the rock - otherwise you will get dates that are all over the place and wouldn't know which date to pick as the correct one. What always bothered me was that when rocks with know ages are dated, dating methods don't work, while rocks with unknown ages are dated, dating methods do work! Now you tell me if you can accurately date anything based on that assumption!

Crusader, thanks for your response. Why do you say you get dates that are all over the place, if the readings are given to the accuracy described above?

 

[Micaiah]

Unsupported claims. Can you provide some details on this? You claim that dating methods do not work when rocks of known ages are dated. This is certainly true when creationist use poor methods of collections such as dating zenoliths within the sample but other than that I think your conclusions are wrong.

Notto,

Can you explain?

 

[notto]

Notto,

Can you explain?

Consistency of dating methods and discussion of xenoliths (and how creationists don't like to talk about them when attacking dating methods).
http://www.gate.net/~rwms/AgeEarth.html

The results from several different methods produce the same results. This cannot be explained by changed in decay rates because the changes would not affect all type in exactly the right way to still give consistent dates (unless of course God changed them specifically to do so, tinkering directly with the decay rate of each element in different proportions all apparently with the goal of fooling us).

Discussion of creationists (purposely) using dating techniques outside of its scope and with poor sample preperation here:
http://www.island.net/~rjbw/CreationScience.html

Here Snelling puposely submitted a rock for dating to a lab that specifically says that their methods would be inaccurate on such a sample. He also requested a 'whole rock' analsysis and did not request that xenoliths be removed. Xenoliths are known to throw of dating (and with good reason). At worst he purposely did this to deceive, at best, he was just sloppy.

 

[Micaiah]

What would you consider a fair test of the method's acccuracy?

Here is a test.

I go out into a patch of rock, dig up three samples and send them off to three different labs. I look at the dates they give and on that basis make a decision about whether the method is any good.

What do you think?

 

[notto]

What would you consider a fair test of the method's acccuracy?

Here is a test.

I go out into a patch of rock, dig up three samples and send them off to three different labs. I look at the dates they give and on that basis make a decision about whether the method is any good.

What do you think?

Well, you would have to document attempts to remove or prevent contamination and xenoliths but it would be a good start. The samples would need to be clean samples of the same rock. Making sure that the right method is applied would be another good step. After your results are back you would need to take into account the tolerance of the method and make sure that you didn't get bad dates due to being at the measurment extremes of the method used. Make sure to avoid the sloppy methods shown by Snelling and don't ignore the labs own statements about what their tests can be used on reliably. Also, don't misrepresent the results by ignoring the discussion of contamination and xenoliths that would accomplany the results.

 

[Micaiah]

Getting lots of caveats here Notto. I want a method that is reliable, not one with more letoffs than a car saleman's warranty.

Try this. I go to a patch of rock for which evolutionists have given a date. I use two different dating methods that they suggest. I take two samples for each test and compare the results. All samples are taken within close proximity.

How close would you expect the results to be in order to say they are reliable. Are we talking within 10 % again. What if we relaxed the criteria for reliability and said 20%.

 

[notto]

Getting lots of caveats here Notto. I want a method that is reliable, not one with more letoffs than a car saleman's warranty.

Try this. I go to a patch of rock for which evolutionists have given a date. I use two different dating methods that they suggest. I take two samples for each test and compare the results. All samples are taken within close proximity.

How close would you expect the results to be in order to say they are reliable. Are we talking within 10 % again. What if we relaxed the criteria for reliability and said 20%.

Well, you need to follow appropriate methods if you want to get appropriate results. That's just the way it works. As with any testing, the right tools, methods, and instruments need to be used. Can't provide you what you are asking for because the methods are only reliable when used correctly.

As far as what I would expect of the results to show that they are reliable, I would probably want a higher sample rate than 3. The more samples you have, the better you can fit a curve to it to get a better approximation of the true result. If the results fall within the +- margin of error of each other, I would consider them reliable. The accuracy might be in question if the margin of error was high.

You are talking about several different statistical concepts.
Accuracy and reliability and the margin of error are based on different things.

 

[Micaiah]

I've been reading an article in an AIG magazine recently about dating using isotopes. They used several different dating methods on rocks from the grand canyon with a known age. They took a number of samples from various locations.

Lets just say the results fall well outside what you and I consider to accurate or reliable.

I'm having trouble understanding the isocrons though. They show ratios of different chemicals on both axes. Maybe you can explain how they work.

 

[notto]

I've been reading an article in an AIG magazine recently about dating using isotopes. They used several different dating methods on rocks from the grand canyon with a known age. They took a number of samples from various locations.

Lets just say the results fall well outside what you and I consider to accurate or reliable.

Well, without looking at the data, I really can't say if they fall well outside of what I would consider to be accurate or reliable. How accurate or reliable was the person who collected the samples? A discussion of that can be found here
http://www.talkorigins.org/faqs/icr-science.html

I'm having trouble understanding the isocrons though. They show ratios of different chemicals on both axes. Maybe you can explain how they work.

A good description and primer of what the graphs mean can be found here.
http://www.talkorigins.org/faqs/isochron-dating.html#isochron

 

[Micaiah]

Thanks Notto. You're a fount of knowledge. Well links anyway. The discussion on isochrons was helpful. The article I'm looking at was by Andrew Snelling fom AIG. Not the one you mention.

The isochrons used were those developed by a computer program called Isoplot.

The material tested was 'Brahma' amphobolites. I understand these are reported to be 1.7 billion years old by many scientists.

 

[AvgJoe]

How reliable are these dating methods ?

Food for thought~~~> http://www.trueorigin.org//dating.asp

 

[Micaiah]

Thanks for the links. The main dfficulty seems to be getting a sample that can be considered 'good'.