DNA testing companies under attack

What theories do you have in mind?

For example that R1b was vasconic cro-magnon hg,
or that N1 are blond inhabitans of Scandinavia since
tens of thousands of years... But such insane things
people were claiming as a scientific facts.

And what exactly do you mean by "the one" just curious.

That R1 = IE.
No vasconic R1b.
No uralo-altaic or dravidian R1a.
 
I do agree with you that genetic science differs from the social science. I studied economics and politcial science and sometimes even (macro) economic models are applicable within political studies. Political science is a social science and even in my field it is necessary to use statistical analyzes to test proposed hypothesis -> thesis, antithesis, synthesis.

Genetic science is much more 'medical' and 'biological'. Without math genetic science is not viable. It is closely related to EXACT science than to social science.

But it is all about patterns if you are using descriptive & inferential statistics and algebra (equations) in general.

Pi (π) is everywehere and where is π there are patterns. It is all about to 'find' those patterns.
 
Those people who are interested in genetic science but refuse to use statistical methology are not (properly) educated. They simply don't understand what they are talking about. It is how science works, at least since Aristotle. It was the FIRST thing what I learned at the university.

And I read here & there that some folks are against 'peer review'. It is one of the fundamental parts of science. Evaluating others maintain the high quality of researchers & studies and it keeps uneducated laymen and pseudoscience out. That keeps science LESS contaminated from people who are uneducated or who have a hidden agenda and those who try to spread lies, propaganda and try to pollute science.


It is all about finding the truth. Our evolution and development is correlated to science. People who are trying to spread lies/propaganda/racism/politics with all their power/money should be banned from the science forever!
 
Humans are very good at seeing patterns and attributing meaning to those patterns even when the patterns have no significance. That's why we have the scientific method which is a way of testing hypotheses.
In this particular discipline of historical population genetics ancient DNA is the final judge, either for scientific models or educated guesses. In a decade or two we should collect so much ancient DNA to figure out all the where how and when they came, went and mixed. Whomever could predict the most, from scientific methods or just from exceptional logic, will be the king. In this field, people with vast knowledge of history and archeology will have huge advantage in connecting the dots.

Humans are very good at seeing patterns and attributing meaning to those patterns even when the patterns have no significance.
Generally yes, but there always are these few geniuses who can see how it is. For the rest of us is the daily grind of methodology to figure out something. And even with right methodologies very often there is a lack of good brain power to pull out the right conclusions. Generally speaking of course as I'm not familiar with your work. Just putting things in perspective.
 
In this particular discipline of historical population genetics ancient DNA is the final judge, either for scientific models or educated guesses. In a decade or two we should collect so much ancient DNA to figure out all the where how and when they came, went and mixed. Whomever could predict the most, from scientific methods or just from exceptional logic, will be the king. In this field, people with vast knowledge of history and archeology will have huge advantage in connecting the dots.

Generally yes, but there always are these few geniuses who can see how it is. For the rest of us is the daily grind of methodology to figure out something. And even with right methodologies very often there is a lack of good brain power to pull out the right conclusions. Generally speaking of course as I'm not familiar with your work. Just putting things in perspective.

Ancient DNA is just one type of evidence. As with any evidence it needs to be used in conjunction with other types of evidence to draw conclusions. Hypotheses still need to be tested using the scientific method. Ancient DNA on its own will never provide the answers. Survival of ancient DNA is patchy. There are places in the world where the climate is not amenable to the survival of ancient DNA (eg, much of Africa). Researching past populations is a huge task and no individual can possibly acquire the necessary knowledge. That is why research is multidisciplinary. We need the population geneticists to do the ancient DNA analysis and the bioinformatics, the historians to put the results in context, and the archaeologists to gather the archaeological evidence. Research is generally more effective when experts from multiple disciplines collaborate.
 
Genetic science is much more 'medical' and 'biological'. Without math genetic science is not viable. It is closely related to EXACT science than to social science.

Historical population genetics has practically nothing medical about it. And anyway, I studied biomedical sciences and I can tell you that maths only play a small part in biology and medicine. And when there is maths, it's mostly about calculating risk factors, probabilities and statistical analysis. Nothing very complex.
 
Historical population genetics has practically nothing medical about it. And anyway, I studied biomedical sciences and I can tell you that maths only play a small part in biology and medicine. And when there is maths, it's mostly about calculating risk factors, probabilities and statistical analysis. Nothing very complex.
What do you mean? DNA is part of our body, it's everywhere. It has to be 'medical'. I don't know anything about this field, but when I think about biomedical science I think immediately about chemistry and chemistry is exact science. To know what your Y-DNA or auDNA is you need chemistry. And chemistry = EXACT science.

Biology is not only what we see, tangtible observation, like human body, bacteria/viruses, flora & founa, but also BEHAVIOR. There is a substream within the economics which we call environmental economy. To understand the behavior of humans we need to study the behavior of animal, fish and birds. Like, about their population balances etc.

In political science behavionarism is also very important. Behavionarism is part of sociologiy and human psychology, but even that field NEEDS statitics for their theories. So, behavionarism can be explained by sociology (history) and economy. And the most scientific way to explain behavionarism is with inferential statistics.

Like I said, I studied also political science. History was part of my edcuation, since we try to study the origin of political theories and paradigms. Most of the times history repeats itself, so you could also use inferential statistics (+economic, environmental and ecological models) if you try to explain history.

So, math (Pi (π) ) is everywhere. Patterns are everywhere, in economy, history, sociology etc. What we need is to find, observe and analyse those patterns. Therefore we need descriptive & inferential statistics.


Every genius in history from Aristotle to Tesla used statistics. Without statistics it is impossible to call yourself a scientist and to be successful..
 
Population geneticists do not study haplogroups per se. The Y-chromosome represents only two per cent of our DNA and, because only males have a Y-chromosome, it only represents half of the human population. mtDNA is a tiny molecule of just 16569 base pairs, and provides a very limited view of our ancestry because it only follows the all-female line. While Y-DNA and mtDNA are very useful for genetic genealogy and can be used very effectively within a genealogical timeframe they are less useful for exploring the history of populations. That is why research now focuses on autosomal DNA which is so much more informative.

I frankly don't understand why you would feel the need to remind me of these things. I did mention in my reply to you above (post #22) that I studied biomedical sciences, including genetics, while you don't seem to have a scientific background (based on your LinkedIn profile). All the things you mention are very, very basic, and I even wrote a page a few years ago with those fundamental concepts of genetics so that I didn't have to re-explain everything to non-biologists on the forum.

It's the second time in this thread that you tell me that mtDNA has 16569 base pairs, even when after the first time I explained that not only I knew it, but I was one of the few people on this planet who had analysed all the mutations in mitochondrial DNA and that I had identified a pattern of mutations in the Coenzyme Q - cytochrome c reductase gene (MT-CYB) defining successful mtDNA lineages and argued that this was the result of natural selection and environmental adaptations. To the best of my knowledge no other geneticist did such a thorough analysis of gene function alterations in mitochondrial haplogroups.

Still in my reply to you in post #22, which you obviously haven't read properly, I explained how I thought that mtDNA was of little use for tracing ancestry other than distant prehistoric one, justly because of the low resolution that mtDNA offers. Here is what I wrote:

As for mtDNA, I have explained for years that it is practically useless for tracing ancestry for at least the last 4000 or 5000 years. At best mtDNA can distinguish between Mesolithic European, Neolithic Near Eastern farmers and Steppe Indo-Europeans, but not anything more since these populations mixed, as mtDNA evolves very slowly and on a 16,569 bases long sequence any mutation can quickly become hazardous for health.

This is something I have been explaining for at least 7 years on this website (one of the largest for historical population genetics), for longer than you have worked in genetic genealogy. So why are you telling me this in a tone that implies that you know better than me or that I surely need to be reminded of the 101 of historical population genetics? At least I had the decency to check your blog and the UCL pages about the BritainsDNA controversy that you mentioned before replying to you. You apparently did not read anything I posted in link, or anything I wrote in the Genetics section of this website (in case you are wondering which articles I wrote, that's all of it). You reply to me as if you had superior knowledge and I was just an amateur, which is ironic considering the extent of your knowledge or your background.


It’s always encouraging to see people coming up with new ideas but any new methods have to be tested to see if they work. You have formulated a hypothesis that Y-DNA and mtDNA haplogroups travel in tandem but how have you tested this hypothesis?

I am yet to see a theoretical or mathematical model in historical population genetics that has achieved any useful prediction. The only way to confirm predictions in this field is to test lots of ancient DNA samples. That's the bottom line. You keep repeating that we won't be able to test samples from every period and every country, but it isn't any more necessary than to test every single individual alive today to know the frequency of haplogroups (or of an allele in a gene, for medical researchers) in a country or region. We will never have perfect knowledge of everything. Even if governments made full genome sequencing compulsory for everyone, with the hundreds of thousands of people who are born and die in the world every day we would need to update frequencies in real time. Not only is that not realistic in our age and time, but it would be quite impractical to manipulate the data, if it keeps changing every millisecond. At the moment, in the field of population genetics, we have to work with the limited amount of samples that is available. Some countries are better sampled and more representative. Others have practically no data. Ancient DNA tests are brand new. The first ancient Y-DNA and autosomal DNA were published only four years ago. But the amount of results is increasing fast and we can expect thousands of ancient samples in the next few years. Eventually we will reach a point when any past theories and predictions about the origins and diffusions of haplogroups and specific prehistoric populations can be verified with a high degree of confidence, one that could very well be higher than the allele frequency databases we have today for modern populations.

If you had read my posts #36 and #37, you'd understand that the scientific method cannot validate any model in a field like historical population genetics, because that field is essentially history (a social science) and genetics is only there to provide data to confirm theories and hypothesis based on archaeological or linguistic evidence. I am a scientist by formation, but I am not going to pretend that there is a way to reduce history to a testable statistical model! That's ludicrous.
 
What do you mean? DNA is part of our body, it's everywhere. It has to be 'medical'. I don't know anything about this field, but if i think about biomedical science I think immediately about chemistryand chemistry is exact science. TO know what your Y-DNA or auDNA is you need chemistry.And chemistry = EXACT science

The DNA itself is exact science. The DNA tests are more or less exact science, because sequencers are not perfect and can misread genetic sequences. That's why good DNA testing companies perform many sequencing (or actually genotyping for most commercial companies) runs, to minimise the number of errors. Sorry to disappoint you, but it's possible that some allele values in your 23andMe test aren't correct. There are surely very few of them, but it's not 100% guaranteed accurate.

Many lay people think that medicine is an exact science, but it is far from being the case. We are only starting to enter (slowly) into the age of precision medicine thanks to DNA tests, not just for our genome, but for the microbiome (bacteria, fungi and viruses) within us and around us. In a near future it should become possible for doctors (and patients) to test almost instantly what kind of pathogen they harbour. That's a huge improvement on how medicine is performed now, which is nothing more than guesswork based on the patient's symptoms. It's recently become possible (and encouraged) to test mutations in cancerous cells so as to be able to determine which treatment could work better. But be it for viruses, bacteria or cancer, treatments are still very random. Not everybody reacts the same way to the same treatment. And microbes can mutate quickly and become resistant today to drugs that worked yesterday. The way new drugs are developed today is also pretty much based on luck, guesswork and a long series of experiments (first on lab animals, then on humans in clinical trials). That's not what I call 'exact science'. In a few decades we will look back in disgust on 20th century and early 21st century medicine as something not much more evolved than witchcraft. Fortunately, that's about to change soon, and it is already changing fast when it comes to cancer treatment thanks to advances in immunotherapy and gene therapy, which is much closer to exact science. But don't ever think that your local GP practices exact science. ;)
 
I don't understand what you are talking about. My background is political science. Like I wrote in my previous post, political science is part of the social science. Scientists with the social science background (sociology, politicology, psychology and EVEN history) have to use descriptive & inferential statistics all the time. Otherwise their studies and theories would never be considered 'scientific'/academic. Social science tries to explain the behavior of humans, political actors, interaction of societies on different levels. To explain behavionarism and explain different (historical) paradigms you have to use statistics.

History doesn't mean only field research or just read a historical book or something like that. It is to UNDERSTAND the history and make future 'predictions'. I know people who studied just history and they are using statistics all the time, since history repeats itself and since there are many patterns in human behavior truth history. That's how you can 'predict' the downfall of dictators (political entities/actors), since there is a correlation (patterns) in their behavior. That's why (and how) you can predict the downfall of let say Turkey.


I don't understand what you are trying to prove. You are 'tilting at windmills'...
 
I don't understand what you are talking about. My background is political science. Like I wrote in my previous post, political science is part of the social science. Scientists with the social science background (sociology, politicology, psychology and EVEN history) have to use descriptive & inferential statistics all the time. Otherwise their studies and theories would never be considered 'scientific'/academic. Social science tries to explain the behavior of humans, political actors, interaction of societies on different levels. To explain behavionarism and explain different (historical) paradigms you have to use statistics.

History doesn't mean only field research or just read a historical book or something like that. It is to UNDERSTAND the history and make future 'predictions'. I know people who studied just history and they are using statistics all the time, since history repeats itself and since there are many patterns in human behavior truth history. That's how you can 'predict' the downfall of dictators (political entities/actors), since there is a correlation (patterns) in their behavior. That's why (and how) you can predict the downfall of let say Turkey.


I don't understand what you are trying to prove. You are 'tilting at windmills'...

You must be kidding, right? There is no serious maths in social sciences. Using statistics, making statistics, making predictions, calculating probabilities and stuff like that aren't real maths. That's the kind of things everybody does. I consider that if something is not harder than university entrance level, then it's not worth mentioning it. I have compiled the statistics table for the haplogroups on this sites, which underlie all my research and my articles and were used to make my haplogroup frequency maps. But I would never say that this is mathematics. If I go to the supermarket and add up mentally the price for all the items in the basket, I am not turning into a mathematician. That's something everybody does. When I translate my articles into French or Italian, I don't consider myself a translator, because that's not my job or my field of research or interest. That's just something I have to do. I made all the web design of this site but you never see me call myself a web designer. I can replace a light fixture or fix a leaky pipe under a sink in my house, but that doesn't make of me an electrician or a plumber. The maths used in social sciences is so basic compared to those used in physics or artificial intelligence that it falls into the same 'everyday use' category. And by the way, I also studied economics and I know that theoretical models in social sciences are far from accurate and sometimes dangerously wrong.

My disagreement with Mark and Debbie above comes mainly from the fact that they seem to believe that it is possible to make useful predictions about historical population migrations solely based on theoretical models, completely disregarding all the other fields I have cited in this discussion. You can predict the trajectory of a comet using theoretical models and maths, but you simply can't do it for haplogroup migrations. I have never seen it work, and I can't imagine how it could possibly work. But I remain open to a demonstration. We seem to hold two completely opposite positions:

- I favour a transdisciplinary approach using logic, statistics and analysis of all evidence from every discipline to maximise our understanding of human population history. I favour hard facts over abstract theoretical models. From my point of view, each large migratory event in history (e.g. Neolithic expansion from the near East) is unique and should be used very carefully to extrapolate on other migrations in different ages, cultures and environments (because humans are not just numbers and their behaviour and success is by nature unpredictable).

- They insist that the only valid and recognised methods in population genetics is based on theoretical models that can be mathematically tested. They believe that once we have found a model that works, it can be applied in any population, like the immutable laws of physics. From their standpoint, my methodology is useless because it is case by case and cannot be tested and re-used (which would be understandable if we were working on exact sciences, but we aren't).


I also believe that there are too many unknown factors even for simple mathematical models to be accurate in that field. For example, the age or TMRCA of haplogroups can be calculated from the accumulated mutations, each tempered by the statistical chance of a mutation occurring at that position in the DNA sequence (since mutations are known to happen more frequently in some places than others). But that doesn't take into account the historical population size at each point in the phylogenetic tree, nor the local radioactivity, which increase the chances of mutation happening. A major difference in population size between two branches of a same haplogroup, for example between Siberian R1a (low population) and Indian R1a (high population) can lead to a number of accumulated mutations hundreds or thousands of times superior in the Indian group over the same period of time. Without knowing exactly where a subclade evolved and what kind of population size it had during this development, calculations are doomed to be mistaken, especially in region with unusually high or low population densities or growth.

In other words, I blame them for having a too simplistic mathematical approach that cannot possibly taken into account all these factors. There has actually been quite a few population geneticists claiming that haplogroup R1a originated in India because genetic diversity was higher there, not realising that this diversity was the result of extremely high historical population size since the Bronze Age. This is what Sharma et al. (2009) claimed, and their study was not only published in a peer reviewed journal, but the most prestigious of them all, Nature. So much for the peer review system filtering out the pseudoscience. That 'peer' word doesn't mean anything. Having graduated in a scientific subject is not a proof of intelligence, critical sense or discernment. In fact the average IQ of university graduates is quite low; 115 to 130 in the US according to this site. Most wouldn't even make it to Mensa.
 
- I favour a transdisciplinary approach using logic, statistics and analysis of all evidence from every discipline to maximise our understanding of human population history. I favour hard facts over abstract theoretical models. From my point of view, each large migratory event in history (e.g. Neolithic expansion from the near East) is unique and should be used very carefully to extrapolate on other migrations in different ages, cultures and environments (because humans are not just numbers and their behaviour and success is by nature unpredictable).

- They insist that the only valid and recognised methods in population genetics is based on theoretical models that can be mathematically tested. They believe that once we have found a model that works, it can be applied in any population, like the immutable laws of physics. From their standpoint, my methodology is useless because it is case by case and cannot be tested and re-used (which would be understandable if we were working on exact sciences, but we aren't).
It is a good description how we all model population history in your brain. We come up with hypothesis, run many models and come up with predictions. Generally not much different than running predictive models in a computer. The difference is that so far good human brain is much better in it, because it analyzes data sets in much broader environment, among others human behavior, religion or climate change.



I also believe that there are too many unknown factors even for simple mathematical models to be accurate in that field. For example, the age or TMRCA of haplogroups can be calculated from the accumulated mutations, each tempered by the statistical chance of a mutation occurring at that position in the DNA sequence (since mutations are known to happen more frequently in some places than others). But that doesn't take into account the historical population size at each point in the phylogenetic tree, nor the local radioactivity, which increase the chances of mutation happening. A major difference in population size between two branches of a same haplogroup, for example between Siberian R1a (low population) and Indian R1a (high population) can lead to a number of accumulated mutations hundreds or thousands of times superior in the Indian group over the same period of time. Without knowing exactly where a subclade evolved and what kind of population size it had during this development, calculations are doomed to be mistaken, especially in region with unusually high or low population densities or growth.
Exactly, and in addition to my previous point. As you mentioned, the complexity of environment is enormes, and it will take a very long time to develop adequate and sophisticated enough computer models to take this huge complexity of human environment under consideration.
 
You must be kidding, right? There is no serious maths in social sciences. Using statistics, making statistics, making predictions, calculating probabilities and stuff like that aren't real maths. That's the kind of things everybody does. I consider that if something is not harder than university entrance level, then it's not worth mentioning it. I have compiled the statistics table for the haplogroups on this sites, which underlie all my research and my articles and were used to make my haplogroup frequency maps. But I would never say that this is mathematics. If I go to the supermarket and add up mentally the price for all the items in the basket, I am not turning into a mathematician. That's something everybody does. When I translate my articles into French or Italian, I don't consider myself a translator, because that's not my job or my field of research or interest. That's just something I have to do. I made all the web design of this site but you never see me call myself a web designer. I can replace a light fixture or fix a leaky pipe under a sink in my house, but that doesn't make of me an electrician or a plumber. The maths used in social sciences is so basic compared to those used in physics or artificial intelligence that it falls into the same 'everyday use' category. And by the way, I also studied economics and I know that theoretical models in social sciences are far from accurate and sometimes dangerously wrong....

...
In other words, I blame them for having a too simplistic mathematical approach that cannot possibly taken into account all these factors. ...

Social science is to broad. Let's take history. I am making it more difficult for me since history is less mathematical than sociology or psychology that makes use of statistics for their theories. They make use of probilities and empirical studies mostly for predicitions. (like very simple statistical null hypothesis testing)


But now let's talk about history and more heavy models.


If you want to measure some population growth within an area truth history you need some kind of environmental population models.

If you want to know how much money French Revolution costed the French society, you need to use some economical models. If you want to know how much money Dutch goverment earned from it's colonies in SouthEast Asia during the Dutch Golden Age you need some heavy models.

How much money did WW1 cost the Germans? For how many years did the development in Europe stopped due to the The Black Death? For all this questions you need different mathematical models.


Scientists are making mistakes all the time(, it is part of a process). But that's because they are using wrong or too simplistic social research methods. But that doesn't mean that using a method is wrong. In contratry in sciense you have to use the models to support your theories. The ART of science is to find the right methology. And that is very difficult. For every problem there has to be a model. Great scienctists can find this model and use it properly, while weak scientists have hard times to find a propriate model or don't know how to use it efficiently..
 
It is a good description how we all model population history in your brain. We come up with hypothesis, run many models and come up with predictions. Generally not much different than running predictive models in a computer. The difference is that so far good human brain is much better in it, because it analyzes data sets in much broader environment, among others human behavior, religion or climate change.

Not to mention that it would take forever to encode all those variables into a computer.

Exactly, and in addition to my previous point. As you mentioned, the complexity of environment is enormes, and it will take a very long time to develop adequate and sophisticated enough computer models to take this huge complexity of human environment under consideration.

It will be much easier and faster once computers/robots will learn to adapt models, measure, gather and encode the data on their own. But at that point humans won't have anything left to do. That should be some time during the 2040's.
 
Social science is to broad. Let's take history. I ammacking it more difficult for me since history is less mathematical thansociology or psychology that makes use of statistics for their theories. Theymake use of probilities and empirical studies mostly for predicitions. (like very simple statistical null hypothesis testing)

But now let talk about history and heavy models.


If you want to measure some population growth withinan area truth history you need some kind of environmental population models.


If you want to know how much money French Revolutioncosted the French society, you need to use some economical models. If you want to know how much money Dutch govermentearned from it's colonies in SouthEast Asia during the Dutch Golden Age youneed some heavy models.


How much money did WW1 cost the Germans? For how manyyears did the development in Europe stopped due to the The Black Death? For allthis questions you need different mathematical models.


Scientists are making mistakes all the time. But that's because they are using wrong or too simplistic social research methods. But that doesn't mean that using a method is wrong. In contratry in sciense you have to use the models to support your theories. The ART of science is to find the right methology. And that is very difficult. For every problem there has to be a model. Great scienctists can find this model and use it properly, while weak scientists have hard times to find a propriate models or don't know how to use it efficently..

In none of these examples can any mathematical model make any useful estimate. There are so many unknown factors and variables that all you'll get is a very, very rough estimate, and often one that is very far from the truth. It is dangerous because people who don't really understand how these calculations are made (most ordinary people) tend to believe the numbers they are fed because it 'looks scientific', even though it's rubbish.
 
In none of these examples can any mathematical model make any useful estimate. There are so many unknown factors and variables that all you'll get is a very, very rough estimate, and often one that is very far from the truth. It is dangerous because people who don't really understand how these calculations are made (most ordinary people) tend to believe the numbers they are fed because it 'looks scientific', even though it's rubbish.
It was just a quick example. Maybe those questions are to broad and need to be rephrased, but I hope that you do understand where I was heading toward. Sometimes you also have to measure correlation (effect or dependence) between let say two historical events, or between events and actors (subject/object) during a given period. Some statistical coefficients are very simple to determine, but sometimes much more work needs to be done to find correlation coefficients.

No, history is not exactly rocket science and according to some folks history is not even science at all because many times it is subsidiary on personal interpretation. But I do believe that history is part of the social science since there are many patterns in it and since I believe that every 'right' question within this discipline can be approached by a statistical model. Historians research & analyze their sources.

more analitical/statistical approach means more scientific process

If my explanation is wrong and you still don't understand me, maybe this is more helpful:


"4. Historians, local historians and statistical analysis" : http://humanities.uwe.ac.uk/bhr/Main/analysis/4-Statistical _analysis.htm
"why historians started counting" : http://historymatters.gmu.edu/mse/numbers/why.html
"Quantitative Methods for Historians" : https://books.google.nl/books/about...r_Historians.html?id=JadX1sfwssQC&redir_esc=y
"QUANTITATIVE SKILLS FOR HISTORIANS" : https://www.heacademy.ac.uk/system/files/rg_freeman_quantitativeskills_20100131_01.pdf
 
Interesting discussion, if I may.

In the humanities or social sciences, research methodology is often viewed as 'soft' (as opposed to hard) science due to a lack of quantifiable data. The majority of papers are qualitative and therefore based on phenomenological and textual analysis.

An important determinant in attempting to acquire 'truth' in a historical context is to determine whether one subscribes to a universal or relativistic truth.

Quantitative analyses as in the field of Mathematics most often provide objective measures for universal 'truth' objectifying and deconstructing constituent variables in meaningful ways. This is the basis of statistical analysis.

However, in qualitative research we are left with subjective meaning units, textual analysis and phenomenological understanding unique to the individual providing the experience/data. In this case one would follow a relativistic 'truth' approach and attempt to extract universal meaning units only if corroborated by quantitative data. This however is contentious as memory has been shown to be neither consistent nor factual.

When we use terms such as 'likelihood' or 'probability' we are using inference to determine a plausible outcome. This is predictive and based on a static contextual environment. Most research questions are based on inference to begin with, the aim being to confirm or deny the hypothesis under review.
 
Interesting discussion, if I may.

In the humanities or social sciences, research methodology is often viewed as 'soft' (as opposed to hard) science due to a lack of quantifiable data. The majority of papers are qualitative and therefore based on phenomenological and textual analysis.

An important determinant in attempting to acquire 'truth' in a historical context is to determine whether one subscribes to a universal or relativistic truth.

Quantitative analyses as in the field of Mathematics most often provide objective measures for universal 'truth' objectifying and deconstructing constituent variables in meaningful ways. This is the basis of statistical analysis.

However, in qualitative research we are left with subjective meaning units, textual analysis and phenomenological understanding unique to the individual providing the experience/data. In this case one would follow a relativistic 'truth' approach and attempt to extract universal meaning units only if corroborated by quantitative data. This however is contentious as memory has been shown to be neither consistent nor factual.

When we use terms such as 'likelihood' or 'probability' we are using inference to determine a plausible outcome. This is predictive and based on a static contextual environment. Most research questions are based on inference to begin with, the aim being to confirm or deny the hypothesis under review.

It is obvious how many times archeologists rewrote prehistory in the last 100 years.

I'm very sceptical about many 'social' studies. I think the science of human psychology will be rewritten many times the next few decades.
 
It was just a quick example. Maybe those questions are to broad and need to be rephrased, but I hope that you do understand where I was heading toward. Sometimes you also have to measure correlation (effect or dependence) between let say two historical events, or between events and actors (subject/object) during a given period. Some statistical coefficients are very simple to determine, but sometimes much more work needs to be done to find correlation coefficients.

No, history is not exactly rocket science and according to some folks history is not even science at all because many times it is subsidiary on personal interpretation. But I do believe that history is part of the social science since there are many patterns in it and since I believe that every 'right' question within this discipline can be approached by a statistical model. Historians research & analyze their sources.

more analitical/statistical approach means more scientific process

If my explanation is wrong and you still don't understand me, maybe this is more helpful:


"4. Historians, local historians and statistical analysis" : http://humanities.uwe.ac.uk/bhr/Main/analysis/4-Statistical _analysis.htm
"why historians started counting" : http://historymatters.gmu.edu/mse/numbers/why.html
"Quantitative Methods for Historians" : https://books.google.nl/books/about...r_Historians.html?id=JadX1sfwssQC&redir_esc=y
"QUANTITATIVE SKILLS FOR HISTORIANS" : https://www.heacademy.ac.uk/system/files/rg_freeman_quantitativeskills_20100131_01.pdf

If I understand well what you are trying to say is that an analytical/statistical approach makes social sciences more scientific? We were talking about the utility of mathematics in social sciences, so I am a bit perplexed by this change of direction since I don't consider data analysis or statistics, nor even statistical analysis to be the prerogative or mathematicians, not any more than the use of language is the prerogative or linguists or philologists. Analytical skills are a prerequisite for all intellectual subjects, and I would go as far as to say that it is indispensable for most aspects of everyday life as well. But then I am surely a far more analytical person than most, and indeed it has happened that people I just met ask me if I am a psychologist or a data analyst because of my tendency to analyse everything around me, including people.

Statistics is just the organisation of data. Analysis is the way the brain process data. Rational thinking wouldn't exist without it. Neither of them is a branch of mathematics in itself, although both are used extensively in mathematics (as in many of subjects). We also use reason, critical thinking, reading and writing in maths, but they aren't branches of mathematics either. That's why I am not really grasping what you are trying to prove. That social sciences are rational endeavours making use of the prefrontal cortex of one's dominant hemisphere? Way to state the obvious.

Nowadays even primary school kids make graphs and pie charts on computers. You can call it 'quantitative skills' if it you prefer the sound of it and it makes it look more professional or important. It's very popular nowadays to try to enhance the image of one's activity using nicer sounding terms, so that a secretary becomes an 'executive assistant' and a cleaner becomes a 'domestic engineer'. That doesn't change their qualifications or job function. It's purely aesthetic (and hypocritical). It's the same with the quantitative skills to describe the use of graphs and statistics in social sciences. That doesn't make it advanced mathematics (i.e. beyond calculus, trigonometry and linear algebra that all secondary schoolers have to learn in order to graduate, at least here in Belgium where there are no elective A-level subjects and everyone must study maths to that minimum level).

Personally I don't care much for linguistic embellissements that make people look and feel more important or intelligent. Many academics who don't have much to say try to hide their incompetence behind flowery jargon. What matters to me is the content, perspicacity and validity of one's arguments, not how many big words one can fill into a sentence. As C.W. Ceran said “Genius is the ability to reduce the complicated to the simple.” This is especially true on a public forum like this one where where one writes for all kinds of people from different backgrounds and with different mother tongues. A good academic or teacher/professor, in my opinion, is someone who can recognise the complexity of one subject, see through that complexity, and re-explain it concisely and clearly for everyone else to understand. As Albert Einstein said (as I am in a quoty mood) “If you can't explain it to a six year old, you don't understand it yourself.” That's why I write my articles about historical population genetics "as simple as possible, but not simpler" (once again getting inspiration from my adolescence idol, Einstein). I think that that approach should be a prerequisite for teaching any subject.
 
Last edited:
Maciamo

I don’t have time right now to get involved in a lengthy discussion so I’m only responding to a few comments very briefly.

As regards cousin matching, everyone has different reasons for testing. I am more interested in using DNA testing to verify my genealogical research.

I’m somewhat confused about your reference to “historical population geneticists”. Historians specialise in history and population geneticists specialise in population genetics. Ideally we should leave the historians to study history and let the population geneticists analyse the getetic data.
And then what? Wait for a reporter to interpret the two and come up with a silly "conclusion" for the sake of good headlines?
This comment screams indifference.
So I just have to give you the benefit of the doubt that had you had more time on your hands, you would have re-read what you have written and not published that part.
 

This thread has been viewed 34455 times.

Back
Top