important resource for people who work in the laboratories such as lua

Previous page 1 23 4 Next page « Previous thread | Next thread »
- lua
- Ex-Muslim
- Grossly Incandescent
- offline
Still Not Significant
Reply #30 - August 03, 2015, 09:39 AM

Quote from: musivore on August 03, 2015, 06:41 AM
Lol.

I'm with you on this one also Q. In every way.

Oh, so then I guess it's Yeezevee and Qtian and Musivore who won't commit. I guess it's true that when you chase two rabbits...or three rabbits...or a whole forum of rabbits, you will not even catch one. :(

At the lab I'm working in right now, even though it's a neuroscience lab, the senior research fellow studied mathematics for most of his career, so he's our statistics guy, I guess. I learned from my old research that it's always a good idea to have someone like that around. Biologists aren't always excellent with numbers...if they were, they would probably have been chemists.
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #31 - August 03, 2015, 11:48 AM

Slightly tangential, but meh.

Mathematics is only part of the statistics tool kit. A mathematician isn't necessarily a good statistician in the same way that a mathematician isn't necessarily a good physicist or economist. There's a fundamental difference between math and stats, math is deductive whilst stats is inductive. Statistics is an extension of epistemology, so philosophy is important here. I contend that much of statistical naivety isn't to do with a lack of mathematical knowledge. Applied statistics (though advanced courses in applied stats can be very mathematical) which is the focus of this thread, isn't extremely math heavy. Mathematical stats on the other hand, is an entirely different beast.

Quote from: Qtian on August 02, 2015, 08:42 AM
An example of the influence of philosophy can be found in the apparent* rift between Bayesians and Frequentists. Those who subscribe to the philosophy of Bayesianism have a different outlook on uncertainty, compared to those who believe in Frequentism. This has an obvious impact on statistics and to the extent that some scientific activity relies on statistical interpretation, it impacts science, too. This is extremely apparent in fields such as machine learning and artificial intelligence. Bayesian methods offer alternatives to null hypothesis significance testing.

Anyways, there are many who think that Bayesian methods are some sort of panacea. If you think that Bayesianism is without philosophical problems, you're sorely mistaken. If you want to do statistics properly then it's probably wise to consider your philosophical assumptions. Statistics and philosophy are inextricably tied together.

*Methods such as Empirical Bayes allow for a fusion of both types of reasoning.

I could go ahead and explain the difference, but here's a pretty cute article : http://simplystatistics.org/2014/10/13/as-an-applied-statistician-i-find-the-frequentists-versus-bayesians-debate-completely-inconsequential/

My mind runs, I can never catch it even if I get a head start.
- yeezevee
- Pub Regular
- offline
Still Not Significant
Reply #32 - August 03, 2015, 12:10 PM

Quote from: Qtian on August 03, 2015, 11:48 AM
Slightly tangential, but meh.

Mathematics is only part of the statistics tool kit. A mathematician isn't necessarily a good statistician in the same way that a mathematician isn't necessarily a good physicist or economist. ....................

Slightly tangential, but meeeeheee....

"Statistics is only part of the Mathematics tool kit. To be a great statistician or a great physicist it is must to have good mathematics background... " ..meeewwwe.

Do not let silence become your legacy.. Question everything
I renounced my faith to become a kafir,
the beloved betrayed me and turned in to a Muslim
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #33 - August 03, 2015, 12:28 PM

Yeez, if you want to actually contribute to the thread then please do contribute.

If you want to post your usual bollocks, please do so elsewhere.

My mind runs, I can never catch it even if I get a head start.
- ibn Bilal
- Global Moderator
- سلاسي هو المسيح
- offline
Still Not Significant
Reply #34 - August 03, 2015, 12:44 PM

Quote from: Quod Sum Eris on May 13, 2015, 04:02 PM
Might as well ask the wind not to blow.
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #35 - August 03, 2015, 12:45 PM

lol

My mind runs, I can never catch it even if I get a head start.
- yeezevee
- Pub Regular
- offline
Still Not Significant
Reply #36 - August 03, 2015, 12:45 PM

Quote from: Qtian on August 03, 2015, 12:28 PM
Yeez, if you want to actually contribute to the thread then please do contribute.

If you want to post your usual bollocks, please do so elsewhere.

Well Qtian bollocks and bullshit stuff in any thread is important to highlight the importance of good stuff such as that link in your OP..

https://mchankins.wordpress.com/2013/04/21/still-not-significant-2/

I agree with that link, very few scientists who works in laboratories have the background of statistics and know the importance of statistics. In fact more than 90% of publications in basic/applied science fields often neglect completely the statistical importance of their results..

Do not let silence become your legacy.. Question everything
I renounced my faith to become a kafir,
the beloved betrayed me and turned in to a Muslim
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #37 - August 03, 2015, 12:49 PM

But in response to you yeez, at no point did I suggest that a good statistician or physicist doesn't need a strong mathematics background. However, that isn't necessarily true of statistics. Take data analysis as an example. It gets by fine without tons of mathematics.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #38 - August 03, 2015, 12:49 PM

Quote from: yeezevee on August 03, 2015, 12:45 PM
Well Qtian bollocks and bullshit stuff in any thread is important to highlight the importance of good stuff such as that link in your OP..

https://mchankins.wordpress.com/2013/04/21/still-not-significant-2/

I agree with that link, very few scientists who works in laboratories have the background of statistics and know the importance of statistics. In fact more than 90% of publications in basic/applied science fields often neglect completely the statistical importance of their results..

I agree. Not sure about the actual percentage of publications, but it's definitely worrying.

Sorry for labeling what you wrote as bollocks.

I'm in a shitty mood today.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #39 - August 03, 2015, 12:52 PM

You guys should check this book out.

(Or you can read the material for free: http://www.statisticsdonewrong.com/)

Quote
Statistics Done Wrong is an illustrative, witty and fun book about the world of statistical reporting in science. It is full of great examples on how data may be misinterpreted due to the lack of understanding of basic statistical concepts and the failure to apply appropriate methodologies. The author is a physicist turned statistician, and as a Ph.D. student in statistics has already accumulated in-depth knowledge and understanding of the most common problems and traps scientists tend to fall into when applying statistical methods. The book highlights the importance of statistical literacy and the current inadequacy of basic statistics education offered to science students in universities. This book is a great read and I would recommend it to any scientist exposed to statistics as well as to any non-scientist who would like to understand more about what may lie behind the ‘statistically significant’ results from studies reported in the media every day.

My mind runs, I can never catch it even if I get a head start.
- yeezevee
- Pub Regular
- offline
Still Not Significant
Reply #40 - August 03, 2015, 12:54 PM

Quote
Quote from: happymurtad on August 03, 2015, 12:44 PM

Quote from: Quod Sum Eris on May 13, 2015, 04:02 PM
Might as well ask the wind not to blow.

Quote from: Qtian on August 03, 2015, 12:45 PM
lol

see that.... that is how the world works...

that posts in that folder were specific TO THE HERO Fred M. Donner.. doctor/actor/professor juicy juice ., and you guys are copying QSE words and putting it here..

Allah says Quran was specific to that time .. and allah says my posts also specific to the folder and responses in it..

Quote from: Qtian on August 03, 2015, 12:52 PM

(Or you can read the material for free: http://www.statisticsdonewrong.com/)

(Clicky for piccy!)

That is very good book specially to those who works in Medical science fields

Do not let silence become your legacy.. Question everything
I renounced my faith to become a kafir,
the beloved betrayed me and turned in to a Muslim
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #41 - August 03, 2015, 01:10 PM

Quote from: lua on August 03, 2015, 09:39 AM
Oh, so then I guess it's Yeezevee and Qtian and Musivore who won't commit. I guess it's true that when you chase two rabbits...or three rabbits...or a whole forum of rabbits, you will not even catch one. :(

You came on too strong, man.

My mind runs, I can never catch it even if I get a head start.
- lua
- Ex-Muslim
- Grossly Incandescent
- offline
Still Not Significant
Reply #42 - August 03, 2015, 01:11 PM

I guess that leaves more time for being the old crone on the mountain.
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #43 - August 03, 2015, 01:16 PM

Will have you have an internet connection in your shack, though?

You're not getting away from CEMB that easily.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
Still Not Significant
Reply #44 - August 03, 2015, 01:17 PM

Also, I'm no rabbit. Musivore said I'm a stallion.

My mind runs, I can never catch it even if I get a head start.
- yeezevee
- Pub Regular
- offline
Still Not Significant
Reply #45 - August 03, 2015, 01:20 PM

Quote from: Qtian on August 03, 2015, 12:49 PM
I agree. Not sure about the actual percentage of publications, but it's definitely worrying.

~~Sorry for labeling what you wrote as bollocks.~~

~~I'm in a shitty mood today.~~

That is all right Qtian., I don't take ANYTHING SERIOUS IN MY LIFE.. ANYTHING ..anything.. I smile at good..bad..ugly .. all such stuff in my life ..

anyways you should make this thread as important resource for people who work in the laboratories such as lua or iua ..lol.. so add some pdf file ....

Quote
How Not to Lie with Statistics
Statistics Done Wrong
The Cult of Statistical Significance

Do not let silence become your legacy.. Question everything
I renounced my faith to become a kafir,
the beloved betrayed me and turned in to a Muslim
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #46 - August 03, 2015, 01:23 PM

Quote from: yeez
anyways you should make this thread as important resource for people who work in the laboratories such as lua or iua ..lol.. so add some pdf file ....

I'm going to rename the thread, now.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #47 - August 03, 2015, 01:54 PM

Quote from: happymurtad on August 02, 2015, 11:54 PM
Reminds me of my student in the UAE who thought the lyric "keep it on the low" was "keep it in the law," since, as he figured, adultery was illegal.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #48 - August 03, 2015, 02:14 PM

Quote from: SCM on August 03, 2015, 08:52 AM
Enjoying this thread. I've been pretty amazed at some of the data bio-type people will try to find a correlation in. I've seen some local biologists present a 'relationship' between two parameters in a scatter plot that looked for all the world like they'd drawn a straight line through a swarm of bees.

Assuming that a higher R^2 automatically translates to a better model is something that I've come across many times.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #49 - August 04, 2015, 12:55 PM

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #50 - August 06, 2015, 03:34 PM

Because Yeez told to me to start recommending books and stuff...

Quote from: http://www.amazon.co.uk/Naked-Statistics-Stripping-Dread-Data-ebook/dp/B007Q6XLF2
Naked Statistics is a good way to remind yourself what statistics is about, or if new to the subject, get a solid grasp of the basics. It is a fine complement to a dry textbook, in that it covers the groundwork in a clear, approachable and entertaining way that is not overly mathematically demanding. Appendices delve deeper into theory and can be read or ignored as the reader wishes.

The first two thirds of the book is particularly good, breezing competently through key statistical concepts up to and including the Central Limit Theorem.

Many people may be drawn to the book because of the growing importance of 'big data'. Wheelan takes this topic on board with a focus on regression analysis, and is not afraid to discuss the pitfalls as well as the benefits of the more abstract 'darker' arts of statistics. However, given the choice between a candid acknowledgements of the fundamental limitations of statistics and an uncomplicated view that 'as long as its done well all will be fine', Wheelan goes in the simpler, more positive direction, even when cheerfully supporting claims that over half of the top-flight peer reviewed scientific papers that draw conclusions from the techniques he proposes are likely to be wrong.

Instead, Wheelan argues that brilliant statistical research simply requires brilliant researchers (guess who?) - and that brilliance is not about being good at the maths, but about a having a creative and intuitive grasp of what works. There are two problems with this. One is that observant readers may well spot flaws in the exemplars Wheelan presents as brilliant. The second (and more important) is that the power of statistics is meant to be its ability to reveal insights that are drawn entirely objectively, yet it is clear that many mistakes in statistical research are due to failings in the researchers' subjective and interpretive skills - in other words, the maths disappears - advanced stats is a matter of judgement (so why not rely on judgement and abandon the somewhat bogus claim of objectivity?).

Consequently (and slightly disappointingly), Wheelan's concluding chapter is all about the amazing contribution statistics will continue to make to solving the world's most pressing problems, rather than a more reflective assessment of its strengths and weaknesses.

All this said, this is a likeable and workmanlike book that treats a potentially dry subject with significant flair.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #51 - August 08, 2015, 10:44 PM

A good book by a good philosopher. Ian Hacking is one of my favourites when it comes to the philosophy of statistics.

The Taming of Chance (Ideas in Context)

Quote from: http://www.amazon.co.uk/The-Taming-Chance-Ideas-Context/dp/0521388848#customerReviews
In this important new study Ian Hacking continues the enquiry into the origins and development of certain characteristic modes of contemporary thought undertaken in such previous works as the best-selling The Emergence of Probability. Professor Hacking shows how by the late nineteenth century it became possible to think of statistical patterns as explanatory in themselves, and to regard the world as not necessarily deterministic in character. In the same period the idea of human nature was displaced by a model of normal people with laws of dispersion. These two parallel transformations fed into each other, so that chance made the world seem less capricious: it was legitimated because it brought order out of chaos. Professor Hacking argues that these developments have led to a new style of scientific reasoning gaining its hold upon us. The greater the level of indeterminism in our conception of the world and of people, the more we expect control and intervention in our lives, and the less we expect freedom. Combining detailed scientific historical research with characteristic philosophic breadth and verve, The Taming of Chance brings out the relations between philosophy, the physical sciences, mathematics and the development of social institutions, and provides a unique and authoritative analysis of the 'probabilisation' of the western world.

My mind runs, I can never catch it even if I get a head start.
- zeca
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #52 - August 10, 2015, 01:51 PM

Quote
Positive result rate for clinical trials dropped from 57% to 8% after requiring registration of studies: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0132382

https://mobile.twitter.com/briannosek/status/629106193773002752

Which side are you on?
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #53 - August 11, 2015, 08:22 PM

Quote from: http://www.theallium.com/math/wanted-statistician-to-make-results-significant/
Wanted: Statistician to make results significant

We are a research laboratory involved in large-scale experimentation.
We have a six-month position available for a creative statistician whose task will be to make all our experimental results significant at P<0.05.
A further six month contract may be offered if you can make the results even more significant.
We need support for our foregone conclusions.
Send your CV, along with 12 letters of reference to prof { @ } theallium.com, quoting REF:IRONYSTAT

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #54 - August 16, 2015, 10:40 PM

Quote from: http://www.amazon.com/The-History-Statistics-Measurement-Uncertainty/dp/067440341X/ref=cm_cr_pr_product_top
This magnificent book is the first comprehensive history of statistics from its beginnings around 1700 to its emergence as a distinct and mature discipline around 1900. Stephen M. Stigler shows how statistics arose from the interplay of mathematical concepts and the needs of several applied sciences including astronomy, geodesy, experimental psychology, genetics, and sociology. He addresses many intriguing questions: How did scientists learn to combine measurements made under different conditions? And how were they led to use probability theory to measure the accuracy of the result? Why were statistical methods used successfully in astronomy long before they began to play a significant role in the social sciences? How could the introduction of least squares predate the discovery of regression by more than eighty years? On what grounds can the major works of men such as Bernoulli, De Moivre, Bayes, Quetelet, and Lexis be considered partial failures, while those of Laplace, Galton, Edgeworth, Pearson, and Yule are counted as successes? How did Galton’s probability machine (the quincunx) provide him with the key to the major advance of the last half of the nineteenth century?

Stigler’s emphasis is upon how, when, and where the methods of probability theory were developed for measuring uncertainty in experimental and observational science, for reducing uncertainty, and as a conceptual framework for quantitative studies in the social sciences. He describes with care the scientific context in which the different methods evolved and identifies the problems (conceptual or mathematical) that retarded the growth of mathematical statistics and the conceptual developments that permitted major breakthroughs.

Statisticians, historians of science, and social and behavioral scientists will gain from this book a deeper understanding of the use of statistical methods and a better grasp of the promise and limitations of such techniques. The product of ten years of research, The History of Statistics will appeal to all who are interested in the humanistic study of science.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #55 - August 16, 2015, 10:42 PM

The Emergence of Probability

Quote from: http://www.amazon.com/Emergence-Probability-Philosophical-Statistical-Probabilistic/dp/0521685575/ref=asap_bc?ie=UTF8
Historical records show that there was no real concept of probability in Europe before the mid-seventeenth century, although the use of dice and other randomizing objects was commonplace. Ian Hacking presents a philosophical critique of early ideas about probability, induction, and statistical inference and the growth of this new family of ideas in the fifteenth, sixteenth, and seventeenth centuries. Hacking invokes a wide intellectual framework involving the growth of science, economics, and the theology of the period. He argues that the transformations that made it possible for probability concepts to emerge have constrained all subsequent development of probability theory and determine the space within which philosophical debate on the subject is still conducted. First published in 1975, this edition includes an introduction that contextualizes his book in light of developing philosophical trends. Ian Hacking is the winner of the Holberg International Memorial Prize 2009.

My mind runs, I can never catch it even if I get a head start.
- gal_from_usa
- Ex-Muslim
- offline
important resource for people who work in the laboratories such as lua
Reply #56 - August 16, 2015, 11:48 PM

Quote from: Qtian on August 16, 2015, 10:42 PM
The Emergence of Probability
(Clicky for piccy!)

Historical records show that there was no real concept of probability in Europe before the mid-seventeenth century, although the use of dice and other randomizing objects was commonplace.

That's very interesting. Does that have any relation to Bible verses like:

Quote
Proverbs 16:33
לג בַּחֵיק, יוּטַל אֶת-הַגּוֹרָל; וּמֵיְהוָה, כָּל-מִשְׁפָּטוֹ. 33
The lot is cast into the lap; but the whole disposing thereof is of the LORD.

that indicate a belief that randomizing things (like lots--basically, you draw from a stack of different sized sticks and the shortest one goes to the person who is chosen--or dice) are not truly random, the results are predetermined by God?

Yea, though I walk through the valley of the shadow of death, I will fear no evil, for I have a sonic screwdriver, a tricorder, and a Type 2 phaser.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #57 - August 17, 2015, 07:43 PM

Quote from: lua on August 03, 2015, 09:39 AM
Biologists aren't always excellent with numbers...if they were, they would probably have been chemists.

In his book 'Naked Statistics', Charles Wheelan makes the claim that the majority of statistical follies aren't due to a lack of numerical ability. I tend to agree with him. Many mishaps occur due to improper reasoning and this isn't something which can be corrected through an increase in mathematical knowledge.

If anyone is interested... watch this video from 12:13 onward:

https://www.youtube.com/watch?v=C42AwvaZ-04

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #58 - August 17, 2015, 08:27 PM

It ain't just scientists who do bad stats, here's an example of a mathematician doing bad statistics:

Quote from: http://www.mathsrevision.net/advanced-level-maths-revision/statistics/continuous-random-variables
A continuous random variable is a random variable where the data can take infinitely many values.

My mind runs, I can never catch it even if I get a head start.
- Qtian
- Pub Regular
- offline
important resource for people who work in the laboratories such as lua
Reply #59 - August 19, 2015, 11:15 PM

Quote from: SCM on August 03, 2015, 08:52 AM
Enjoying this thread. I've been pretty amazed at some of the data bio-type people will try to find a correlation in. I've seen some local biologists present a 'relationship' between two parameters in a scatter plot that looked for all the world like they'd drawn a straight line through a swarm of bees.

This was funny. It might be because I'm currently intoxicated, but I thought that I'd let you know.

Edit: Sober now. Still funny.

My mind runs, I can never catch it even if I get a head start.
Previous page 1 23 4 Next page « Previous thread | Next thread »