-
The Cycling News forum is looking to add some volunteer moderators with Red Rick's recent retirement. If you're interested in helping keep our discussions on track, send a direct message to
In the meanwhile, please use the Report option if you see a post that doesn't fit within the forum rules.
Thanks!
Re: Re:
Since the chart is for a power output of 250 watts, yes I'm sure that a pro-level cyclist COULD produce more power on BOTH the upstroke and the downstroke. But the question is whether he would choose to do that.
My assumption/guess/hypothesis is that the chart shows upstroke & downstroke power levels that the cyclist felt was comfortable and efficient to produce the desired 250 watts.
If a higher power level was desired, I'm certain that more downstoke power would be used, but I'm less certain about
how/if the upstroke power would change. I doubt the upstroke would be allowed to decrease to produce negative torque (unless the cyclist was extremely fatigued).
Jay Kosta
Endwell NY USA
-----backdoor said:
Since the chart is for a power output of 250 watts, yes I'm sure that a pro-level cyclist COULD produce more power on BOTH the upstroke and the downstroke. But the question is whether he would choose to do that.
My assumption/guess/hypothesis is that the chart shows upstroke & downstroke power levels that the cyclist felt was comfortable and efficient to produce the desired 250 watts.
If a higher power level was desired, I'm certain that more downstoke power would be used, but I'm less certain about
how/if the upstroke power would change. I doubt the upstroke would be allowed to decrease to produce negative torque (unless the cyclist was extremely fatigued).
Jay Kosta
Endwell NY USA
Re: Re:
Is this man using "circular" or "mashing",
https://www.youtube.com/watch?v=7-J2BwP6bng
JayKosta said:
Is this man using "circular" or "mashing",
https://www.youtube.com/watch?v=7-J2BwP6bng
The chart is made up. Frank is quite handy with MS Paint.
This man is pedalling circular. Try single leg pedalling without a counter weight.
He is also fricking awesome and inspiring!
This man is pedalling circular. Try single leg pedalling without a counter weight.
He is also fricking awesome and inspiring!
Re:
On the climb he is using the circular style but on the level roads he appears to be using a combination of circular and mashing, where like a masher his torque application is clearly concentrated around 2 - 4 o'c and then doing just enough to bring the pedal back, up and over TDC.
CoachFergie said:
On the climb he is using the circular style but on the level roads he appears to be using a combination of circular and mashing, where like a masher his torque application is clearly concentrated around 2 - 4 o'c and then doing just enough to bring the pedal back, up and over TDC.
Re:
That's not surprising because pedalling technique is something you ignore in all riders. It would not be possible for a two legged circular rider or powercranker to apply power as he does because unlike these two legged riders, his brain is not distracted by what it should be getting the other leg to do during its upstroke.
CoachFergie said:
That's not surprising because pedalling technique is something you ignore in all riders. It would not be possible for a two legged circular rider or powercranker to apply power as he does because unlike these two legged riders, his brain is not distracted by what it should be getting the other leg to do during its upstroke.
Extraordinary claims require extraordinary evidence. You have supplied nothing but claims.
I watched a Schools TT with a range of abilities from newbies to a young girl I coach who is headed to her second Junior World Champs. Some awful positions, some bad gear selections, some were riding in toe clips and straps and one girl forgot her cycling shoes and was riding in school shoes on clipless pedals. If you just looked at pedalling you would see no difference between all of them.
I watched a Schools TT with a range of abilities from newbies to a young girl I coach who is headed to her second Junior World Champs. Some awful positions, some bad gear selections, some were riding in toe clips and straps and one girl forgot her cycling shoes and was riding in school shoes on clipless pedals. If you just looked at pedalling you would see no difference between all of them.
Re:
Of course they were all using the same technique, the natural method they soon discovered as a child on their first bike/trike. Technique will only change when a conscious effort is made to do so. The brain has to be given the new objectives to be followed when putting the muscles to work.
CoachFergie said:
Of course they were all using the same technique, the natural method they soon discovered as a child on their first bike/trike. Technique will only change when a conscious effort is made to do so. The brain has to be given the new objectives to be followed when putting the muscles to work.
Since none of the self-proclaimed experts have decided to come back and explain how to "properly" interpret a p-value I am going to give a little primer. Here is how it works.
Anytime someone does a study comparing two things one wants to know if there is a true difference between those things. The problem is that whenever one measures things there is a lot of random variation in the data such that if there is a difference in the data how does one know if that difference is due to random variation or a real difference between the groups. This is why the science of statistics was developed, to look at data and help one to determine whether the differences seen is due to randomness or something real. To do this statistics generally looks at what is called the "null hypothesis" which is simply the expectation that there is "no difference (a null) between the groups. The null hypothesis is always the same, that it is hypothesized there is no difference between the groups being compared. This is what the p-value represents, the probability that the null hypothesis (no difference between the groups) is true. A study never "proves" anything but can only give you a probability that the results you see does or doesn't represent a real difference between groups. If I want to know if a coin is fair and I throw it 1000 times and come up with 1000 heads the probability is very small that the coin is fair but it isn't proven because it is possible for random variation to give 1000 heads right out of the gate.
In science, many journals have come to require authors to have a probability cut-off of a p<0.05 (a less than 1 in 20 chance that the data difference seen is due to chance) before one can say that the difference is "significant". To these journals, a 1 in 19 chance the data represents a real difference fails to reach their arbitrary cut-off and the author is forces to right that statistical significance was not reached. Oh, and this "failure" says nothing about why the study failed to show a difference to this level (were the numbers too small, the study poorly designed, the evaluation period too short, etc.) or if a difference was shown as to the mechanism to explain the difference. All the number says is the probability that any difference seen in the data is due to chance or represents a real difference between groups.
This is a problem in all sorts of things when it comes to studies that we are interested in. Here are a couple of examples. A well discussed crank length study by Dr. Martin ended up with this data:
Now, the data shows that the differences between both 145 and 170 when compared to both 120 and 220 reach the 95% confidence level that the difference seen represents a real difference and not due to chance. But, this data also suggests that there might be a real difference between 145 and 170 but the data did not reach the 95% confidence level so the author is "forced" to say there is no difference between the two crank lengths. Yet, what we should really be asking is what was the p-value for the difference seen? Of course, he doesn't give it to us so we are forced to wonder. Was there only a 10% chance there is no difference between these crank lengths? But, Martin didn't give us the raw data nor this p-value so all we have is his assessment that there is no difference between these groups when, I think, it is clear there probably is. Why didn't someone look at this and say, "This study needs to be repeated with more people or a better design so we can know for sure." But, instead, everyone accepts the "fact" that because the difference seen didn't reach the p<0.05 level that this means it is proven there is no difference. Nothing is further from the truth.
Earlier, I posted a study that did give raw p values in a PowerCranks study. The difference between the groups did not reach the 95% confidence level but differences between groups when looking at power and efficiency did reach the 87.5% and 75% confidence level. With such a result is it silly to say that failure to reach p<0.05 "proves" PowerCranks don't work (yet Fergie seems to have no trouble doing so). Instead, one should be asking was there something about the study design (not enough participants, not long enough, insufficient intervention, etc.) that might result in another result should the study be redesigned and redone. The purpose of science is to find the truth, not to reinforce ones current bias. Looking at actual p-values when they don't reach the 0.05 cut-off level can help one determine whether a study should be redesigned and redone in a quest for new knowledge. Even when they do reach the 0.05 level it is useful to know if the p is 0.04 or 0.0001. If a study fails to give actual p-values (especially when they fail to reach the 0.05 cut-off) is does a disservice to those reading the study and interested in the topic and the advancement of science.
Anytime someone does a study comparing two things one wants to know if there is a true difference between those things. The problem is that whenever one measures things there is a lot of random variation in the data such that if there is a difference in the data how does one know if that difference is due to random variation or a real difference between the groups. This is why the science of statistics was developed, to look at data and help one to determine whether the differences seen is due to randomness or something real. To do this statistics generally looks at what is called the "null hypothesis" which is simply the expectation that there is "no difference (a null) between the groups. The null hypothesis is always the same, that it is hypothesized there is no difference between the groups being compared. This is what the p-value represents, the probability that the null hypothesis (no difference between the groups) is true. A study never "proves" anything but can only give you a probability that the results you see does or doesn't represent a real difference between groups. If I want to know if a coin is fair and I throw it 1000 times and come up with 1000 heads the probability is very small that the coin is fair but it isn't proven because it is possible for random variation to give 1000 heads right out of the gate.
In science, many journals have come to require authors to have a probability cut-off of a p<0.05 (a less than 1 in 20 chance that the data difference seen is due to chance) before one can say that the difference is "significant". To these journals, a 1 in 19 chance the data represents a real difference fails to reach their arbitrary cut-off and the author is forces to right that statistical significance was not reached. Oh, and this "failure" says nothing about why the study failed to show a difference to this level (were the numbers too small, the study poorly designed, the evaluation period too short, etc.) or if a difference was shown as to the mechanism to explain the difference. All the number says is the probability that any difference seen in the data is due to chance or represents a real difference between groups.
This is a problem in all sorts of things when it comes to studies that we are interested in. Here are a couple of examples. A well discussed crank length study by Dr. Martin ended up with this data:
Earlier, I posted a study that did give raw p values in a PowerCranks study. The difference between the groups did not reach the 95% confidence level but differences between groups when looking at power and efficiency did reach the 87.5% and 75% confidence level. With such a result is it silly to say that failure to reach p<0.05 "proves" PowerCranks don't work (yet Fergie seems to have no trouble doing so). Instead, one should be asking was there something about the study design (not enough participants, not long enough, insufficient intervention, etc.) that might result in another result should the study be redesigned and redone. The purpose of science is to find the truth, not to reinforce ones current bias. Looking at actual p-values when they don't reach the 0.05 cut-off level can help one determine whether a study should be redesigned and redone in a quest for new knowledge. Even when they do reach the 0.05 level it is useful to know if the p is 0.04 or 0.0001. If a study fails to give actual p-values (especially when they fail to reach the 0.05 cut-off) is does a disservice to those reading the study and interested in the topic and the advancement of science.
Seriously? First post back and this *** starts again?
"The p-value is not the probability that the null hypothesis is true or the probability that the alternative hypothesis is false. It is not connected to either. In fact, frequentist statistics does not and cannot attach probabilities to hypotheses."
I'm not even going to give the reference, you've had it enough times and were the first one to post it.
Edit:
I'm putting the post in in case it is edited:
"The p-value is not the probability that the null hypothesis is true or the probability that the alternative hypothesis is false. It is not connected to either. In fact, frequentist statistics does not and cannot attach probabilities to hypotheses."
I'm not even going to give the reference, you've had it enough times and were the first one to post it.
Edit:
I'm putting the post in in case it is edited:
FrankDay said:Since none of the self-proclaimed experts have decided to come back and explain how to "properly" interpret a p-value I am going to give a little primer. Here is how it works.
Anytime someone does a study comparing two things one wants to know if there is a true difference between those things. The problem is that whenever one measures things there is a lot of random variation in the data such that if there is a difference in the data how does one know if that difference is due to random variation or a real difference between the groups. This is why the science of statistics was developed, to look at data and help one to determine whether the differences seen is due to randomness or something real. To do this statistics generally looks at what is called the "null hypothesis" which is simply the expectation that there is "no difference (a null) between the groups. The null hypothesis is always the same, that it is hypothesized there is no difference between the groups being compared. This is what the p-value represents, the probability that the null hypothesis (no difference between the groups) is true. A study never "proves" anything but can only give you a probability that the results you see does or doesn't represent a real difference between groups. If I want to know if a coin is fair and I throw it 1000 times and come up with 1000 heads the probability is very small that the coin is fair but it isn't proven because it is possible for random variation to give 1000 heads right out of the gate.
In science, many journals have come to require authors to have a probability cut-off of a p<0.05 (a less than 1 in 20 chance that the data difference seen is due to chance) before one can say that the difference is "significant". To these journals, a 1 in 19 chance the data represents a real difference fails to reach their arbitrary cut-off and the author is forces to right that statistical significance was not reached. Oh, and this "failure" says nothing about why the study failed to show a difference to this level (were the numbers too small, the study poorly designed, the evaluation period too short, etc.) or if a difference was shown as to the mechanism to explain the difference. All the number says is the probability that any difference seen in the data is due to chance or represents a real difference between groups.
This is a problem in all sorts of things when it comes to studies that we are interested in. Here are a couple of examples. A well discussed crank length study by Dr. Martin ended up with this data:Now, the data shows that the differences between both 145 and 170 when compared to both 120 and 220 reach the 95% confidence level that the difference seen represents a real difference and not due to chance. But, this data also suggests that there might be a real difference between 145 and 170 but the data did not reach the 95% confidence level so the author is "forced" to say there is no difference between the two crank lengths. Yet, what we should really be asking is what was the p-value for the difference seen? Of course, he doesn't give it to us so we are forced to wonder. Was there only a 10% chance there is no difference between these crank lengths? But, Martin didn't give us the raw data nor this p-value so all we have is his assessment that there is no difference between these groups when, I think, it is clear there probably is. Why didn't someone look at this and say, "This study needs to be repeated with more people or a better design so we can know for sure." But, instead, everyone accepts the "fact" that because the difference seen didn't reach the p<0.05 level that this means it is proven there is no difference. Nothing is further from the truth.
Earlier, I posted a study that did give raw p values in a PowerCranks study. The difference between the groups did not reach the 95% confidence level but differences between groups when looking at power and efficiency did reach the 87.5% and 75% confidence level. With such a result is it silly to say that failure to reach p<0.05 "proves" PowerCranks don't work (yet Fergie seems to have no trouble doing so). Instead, one should be asking was there something about the study design (not enough participants, not long enough, insufficient intervention, etc.) that might result in another result should the study be redesigned and redone. The purpose of science is to find the truth, not to reinforce ones current bias. Looking at actual p-values when they don't reach the 0.05 cut-off level can help one determine whether a study should be redesigned and redone in a quest for new knowledge. Even when they do reach the 0.05 level it is useful to know if the p is 0.04 or 0.0001. If a study fails to give actual p-values (especially when they fail to reach the 0.05 cut-off) is does a disservice to those reading the study and interested in the topic and the advancement of science.
Re:
I don't think I've seen anyone on sports forums apply distortion, cherry picking, confirmation bias to scientific study and a whole gamut of logical fallacies as much as you have consistently done for year after year after year.
Frank, this has been explained several times already in this thread. The problem is (once again) your complete inability to understand and/or your desire to deliberately mislead. Which is it?FrankDay said:
That's pretty funny coming from you.FrankDay said:
I don't think I've seen anyone on sports forums apply distortion, cherry picking, confirmation bias to scientific study and a whole gamut of logical fallacies as much as you have consistently done for year after year after year.
First post back after his third ban on this forum and FD is back at it, trying to derail a thread.
Let's not forget the original post which shows a rider with 7 years of immersion training in pedalling in a circular fashion improved efficiency by changing to a more mashing style of pedalling.
Let's not forget the original post which shows a rider with 7 years of immersion training in pedalling in a circular fashion improved efficiency by changing to a more mashing style of pedalling.
Re:
"The P-value is often incorrectly interpreted as the probability that the null hypothesis is
true. Try not to make this mistake. In a frequentist interpretation of probability, there is
nothing random about whether the hypothesis is true, the randomness is in the process
generating the data. One can interpret “the probability that the null hypothesis is true” using
subjective probability, a measure of one’s belief that the null hypothesis is true. One can
then calculate this subjective probability by specifying a prior probability (subjective belief
before looking at the data) that the null hypothesis is true, and then use the data and the
model to update one’s subjective probability. This is called the Bayesian approach because
Bayes’ Theorem is used to update subjective probabilities to reflect new information."
Taken from: http://www.stat.ualberta.ca/~hooper/teaching/misc/Pvalue.pdf
FrankDay said:
"The P-value is often incorrectly interpreted as the probability that the null hypothesis is
true. Try not to make this mistake. In a frequentist interpretation of probability, there is
nothing random about whether the hypothesis is true, the randomness is in the process
generating the data. One can interpret “the probability that the null hypothesis is true” using
subjective probability, a measure of one’s belief that the null hypothesis is true. One can
then calculate this subjective probability by specifying a prior probability (subjective belief
before looking at the data) that the null hypothesis is true, and then use the data and the
model to update one’s subjective probability. This is called the Bayesian approach because
Bayes’ Theorem is used to update subjective probabilities to reflect new information."
Taken from: http://www.stat.ualberta.ca/~hooper/teaching/misc/Pvalue.pdf
Re:
So, given the two examples I used how should the lay person reading this thread interpret those to data results. In one instance two groups are compared and the difference in power and efficiency change between the two groups reached the 0.125 and 0.25 level. What should that mean to the reader?
Then, in the crank length study what does it mean (in real world terms) that the 145 and 170 crank lengths reached the p<0.05 level compared to 120 and 195 cranks? And, in view of there being an obvious difference between 145 and 170 mm cranks there must be a p value comparing those differences. If Martin had given us that value how should that be interpreted as regards any difference in crank length in power generation as determined by this study?
By the way, I agree that the p doesn't represent the probability that the null hypothesis is true. In fact, it is unlikely that the null hypothesis is ever true since the null hypothesis is the two groups are the same and this would never be the same. For the null hypothesis to be the same both groups would have to have the exact same result and the exact same standard deviation. I guess that is possible but the likelyhood is tiny, tiny, tiny. However, that is how it is generally interpreted as these studies are designed to determine if a certain thing has an affect. If the differences are small it is reasonable to assume the groups are the same. But, you are right, it is not technically correct. Good to point out in your statistics class but perhaps a bit over the top for this forum. That is why we are relying on you to tell those of us who don't have your background how to properly interpret and talk about this kind of data.
LOL. Welcome back King, I thought you might have dropped off the end of the earth since you didn't respond to my challenge to come here and educate us as to how to properly interpret a p-value given a real world example. Of course, I knew that to not be the case since you were posting on other threads. I interpreted your failure simply to represent you didn't want to post "the truth" because it effectively back what I am saying. But, alas, when you do return, all you do is point out that a small aspect of what I said is wrong without saying what would be more proper to say or to address the original question to you. Anyhow, here is another chance.King Boonen said:Seriously? First post back and this **** starts again?
"The p-value is not the probability that the null hypothesis is true or the probability that the alternative hypothesis is false. It is not connected to either. In fact, frequentist statistics does not and cannot attach probabilities to hypotheses."
I'm not even going to give the reference, you've had it enough times and were the first one to post it.
Edit:
I'm putting the post in in case it is edited:
FrankDay said:Since none of the self-proclaimed experts have decided to come back and explain how to "properly" interpret a p-value I am going to give a little primer. Here is how it works.
Anytime someone does a study comparing two things one wants to know if there is a true difference between those things. The problem is that whenever one measures things there is a lot of random variation in the data such that if there is a difference in the data how does one know if that difference is due to random variation or a real difference between the groups. This is why the science of statistics was developed, to look at data and help one to determine whether the differences seen is due to randomness or something real. To do this statistics generally looks at what is called the "null hypothesis" which is simply the expectation that there is "no difference (a null) between the groups. The null hypothesis is always the same, that it is hypothesized there is no difference between the groups being compared. This is what the p-value represents, the probability that the null hypothesis (no difference between the groups) is true. A study never "proves" anything but can only give you a probability that the results you see does or doesn't represent a real difference between groups. If I want to know if a coin is fair and I throw it 1000 times and come up with 1000 heads the probability is very small that the coin is fair but it isn't proven because it is possible for random variation to give 1000 heads right out of the gate.
In science, many journals have come to require authors to have a probability cut-off of a p<0.05 (a less than 1 in 20 chance that the data difference seen is due to chance) before one can say that the difference is "significant". To these journals, a 1 in 19 chance the data represents a real difference fails to reach their arbitrary cut-off and the author is forces to right that statistical significance was not reached. Oh, and this "failure" says nothing about why the study failed to show a difference to this level (were the numbers too small, the study poorly designed, the evaluation period too short, etc.) or if a difference was shown as to the mechanism to explain the difference. All the number says is the probability that any difference seen in the data is due to chance or represents a real difference between groups.
This is a problem in all sorts of things when it comes to studies that we are interested in. Here are a couple of examples. A well discussed crank length study by Dr. Martin ended up with this data:Now, the data shows that the differences between both 145 and 170 when compared to both 120 and 220 reach the 95% confidence level that the difference seen represents a real difference and not due to chance. But, this data also suggests that there might be a real difference between 145 and 170 but the data did not reach the 95% confidence level so the author is "forced" to say there is no difference between the two crank lengths. Yet, what we should really be asking is what was the p-value for the difference seen? Of course, he doesn't give it to us so we are forced to wonder. Was there only a 10% chance there is no difference between these crank lengths? But, Martin didn't give us the raw data nor this p-value so all we have is his assessment that there is no difference between these groups when, I think, it is clear there probably is. Why didn't someone look at this and say, "This study needs to be repeated with more people or a better design so we can know for sure." But, instead, everyone accepts the "fact" that because the difference seen didn't reach the p<0.05 level that this means it is proven there is no difference. Nothing is further from the truth.
Earlier, I posted a study that did give raw p values in a PowerCranks study. The difference between the groups did not reach the 95% confidence level but differences between groups when looking at power and efficiency did reach the 87.5% and 75% confidence level. With such a result is it silly to say that failure to reach p<0.05 "proves" PowerCranks don't work (yet Fergie seems to have no trouble doing so). Instead, one should be asking was there something about the study design (not enough participants, not long enough, insufficient intervention, etc.) that might result in another result should the study be redesigned and redone. The purpose of science is to find the truth, not to reinforce ones current bias. Looking at actual p-values when they don't reach the 0.05 cut-off level can help one determine whether a study should be redesigned and redone in a quest for new knowledge. Even when they do reach the 0.05 level it is useful to know if the p is 0.04 or 0.0001. If a study fails to give actual p-values (especially when they fail to reach the 0.05 cut-off) is does a disservice to those reading the study and interested in the topic and the advancement of science.
So, given the two examples I used how should the lay person reading this thread interpret those to data results. In one instance two groups are compared and the difference in power and efficiency change between the two groups reached the 0.125 and 0.25 level. What should that mean to the reader?
Then, in the crank length study what does it mean (in real world terms) that the 145 and 170 crank lengths reached the p<0.05 level compared to 120 and 195 cranks? And, in view of there being an obvious difference between 145 and 170 mm cranks there must be a p value comparing those differences. If Martin had given us that value how should that be interpreted as regards any difference in crank length in power generation as determined by this study?
By the way, I agree that the p doesn't represent the probability that the null hypothesis is true. In fact, it is unlikely that the null hypothesis is ever true since the null hypothesis is the two groups are the same and this would never be the same. For the null hypothesis to be the same both groups would have to have the exact same result and the exact same standard deviation. I guess that is possible but the likelyhood is tiny, tiny, tiny. However, that is how it is generally interpreted as these studies are designed to determine if a certain thing has an affect. If the differences are small it is reasonable to assume the groups are the same. But, you are right, it is not technically correct. Good to point out in your statistics class but perhaps a bit over the top for this forum. That is why we are relying on you to tell those of us who don't have your background how to properly interpret and talk about this kind of data.
Re: Re:
All I have asked is for you (or anyone) to point me to the post where this is explained, especially as regards the real world data I put forth. I can't find what you refer to so I have to rely on you to point me to the post. Telling somebody they are wrong without telling them what is correct is a form of academic bullying (IMHO). You guys seem pretty good at that.Alex Simmons/RST said:
Re: Re:
No it is not a small aspect, it is the basis of your whole argument that is wrong. I didn't respond to your question because it has absolutely nothing to do with what I was telling you, that p-values cannot be used to assign a probability.
There is no such thing as the 0.125 or 0.25 "level". A p-value for significance is chosen BEFORE any stats are done, not after. I'm assuming the authors chose 0.05? The only STATISTICAL interpretation of those number is that they are not statistically different assuming a 0.05 level, if it was chosen BEFORE analysis. This is just another example of how you either have no idea about the statistical interpretation or are purposefully attempting to cherry-pick things you wrongly think enforce your argument due to your complete misunderstanding of the statistics.
I do not care how you interpret that data. You can say you think they are the most brilliant data in the world. what you cannot do is try and use a p-value to assign a percentage of likely significance. You can discuss the design, the real numbers, why you think they are meaningful but you cannot falsely attempt to back it up with a misinterpretation of the statistics.
In general terms you seriously expect someone to assess a whole experiment based on two numbers without even giving the relevant constraints? That is ridiculous. They can ONLY assess it based on whatever level was chosen and cannot offer any other insight into the data.
Firstly, it is 120 and 220, not 120 and 195. It's right there in the graph you posted.
Secondly, I don't care, again this has nothing to do with our previous discussion and I would hazard this thread at all. How was the p-value calculated? What was compared to what? ANOVA? T-test? Were they corrected? If it's a t-test your whole question is wrong as it cannot compare two populations against two other populations. Were the populations combined? How does this relate to the initial hypothesis? Why must there be a p-value comparing 145 and 170? Was it calculated? Why are they obviously different? The study appears to only care about power from what little you presented, so they can only be assessed in such terms. If the power is not statistically different, in terms of this study, statistically, the cranks are not different either. Just like if I painted a set of cranks red. They would obviously be different to black ones, but not in terms of this experiment if power output is the same. Even if you can reject the null hypothesis, this does not mean that the alternative hypothesis is true because the statistics have no idea what the hypotheses are, only those educated in the field can decide that. The very people who are disagreeing with you (Not me. I am an analytical biochemist and I have only been pointing out your misappropriation of p-values).
Don't answer those questions, I've already pointed out I don't care and it has nothing to do with our previous discussion. I'm just pointing out how silly it is to expect anyone to interpret a whole experiment based on one number (not even a number) and a supposition.
No, this is again wrong. The null hypothesis is that there is no statistical difference between the groups, not that they are exactly the same, it is the result is due to chance alone. The observed populations can vary and are expected to, there would be absolutely no need to calculate a p-value if your hypothesis were these are different, you'd just look at the numbers. The whole point of a p-value is to give some rigor in how those numbers are interpreted BECAUSE of that variation.
It is not a small point, and it is not how they are interpreted. It is completely fundamental to any interpretation of a p-value and seems to be the basis of your misunderstanding. The populations can vary wildly, the null hypothesis is that any variation is due to chance alone.
You're not relying on me at all. If you were then this conversation would have lasted for about three posts. Your initial post, mine telling you that you cannot interpret a p-value in that way and possibly your response saying thanks (I say possibly because I wouldn't have expected it, this is not a personal dig). That would have been it. Instead you have constantly mis-interpreted references, at times falsely edited them to fit your argument and tried to engage me in discussions I do not want to have and that have no relevance to the point being discussed.
My ONLY interest in this whole conversation is that you stop falsely using p-values. Other than that I do not care how you interpret the data, I am no way qualified to assess that.
The people you should be relying on for that are the coaches and exercise physiologists. Only they have the relevant experience and knowledge to interpret the data beyond the statistics and explain why it is happening. But, unfortunately, they don't agree with you.
FrankDay said:
No it is not a small aspect, it is the basis of your whole argument that is wrong. I didn't respond to your question because it has absolutely nothing to do with what I was telling you, that p-values cannot be used to assign a probability.
There is no such thing as the 0.125 or 0.25 "level". A p-value for significance is chosen BEFORE any stats are done, not after. I'm assuming the authors chose 0.05? The only STATISTICAL interpretation of those number is that they are not statistically different assuming a 0.05 level, if it was chosen BEFORE analysis. This is just another example of how you either have no idea about the statistical interpretation or are purposefully attempting to cherry-pick things you wrongly think enforce your argument due to your complete misunderstanding of the statistics.
I do not care how you interpret that data. You can say you think they are the most brilliant data in the world. what you cannot do is try and use a p-value to assign a percentage of likely significance. You can discuss the design, the real numbers, why you think they are meaningful but you cannot falsely attempt to back it up with a misinterpretation of the statistics.
In general terms you seriously expect someone to assess a whole experiment based on two numbers without even giving the relevant constraints? That is ridiculous. They can ONLY assess it based on whatever level was chosen and cannot offer any other insight into the data.
Firstly, it is 120 and 220, not 120 and 195. It's right there in the graph you posted.
Secondly, I don't care, again this has nothing to do with our previous discussion and I would hazard this thread at all. How was the p-value calculated? What was compared to what? ANOVA? T-test? Were they corrected? If it's a t-test your whole question is wrong as it cannot compare two populations against two other populations. Were the populations combined? How does this relate to the initial hypothesis? Why must there be a p-value comparing 145 and 170? Was it calculated? Why are they obviously different? The study appears to only care about power from what little you presented, so they can only be assessed in such terms. If the power is not statistically different, in terms of this study, statistically, the cranks are not different either. Just like if I painted a set of cranks red. They would obviously be different to black ones, but not in terms of this experiment if power output is the same. Even if you can reject the null hypothesis, this does not mean that the alternative hypothesis is true because the statistics have no idea what the hypotheses are, only those educated in the field can decide that. The very people who are disagreeing with you (Not me. I am an analytical biochemist and I have only been pointing out your misappropriation of p-values).
Don't answer those questions, I've already pointed out I don't care and it has nothing to do with our previous discussion. I'm just pointing out how silly it is to expect anyone to interpret a whole experiment based on one number (not even a number) and a supposition.
No, this is again wrong. The null hypothesis is that there is no statistical difference between the groups, not that they are exactly the same, it is the result is due to chance alone. The observed populations can vary and are expected to, there would be absolutely no need to calculate a p-value if your hypothesis were these are different, you'd just look at the numbers. The whole point of a p-value is to give some rigor in how those numbers are interpreted BECAUSE of that variation.
It is not a small point, and it is not how they are interpreted. It is completely fundamental to any interpretation of a p-value and seems to be the basis of your misunderstanding. The populations can vary wildly, the null hypothesis is that any variation is due to chance alone.
You're not relying on me at all. If you were then this conversation would have lasted for about three posts. Your initial post, mine telling you that you cannot interpret a p-value in that way and possibly your response saying thanks (I say possibly because I wouldn't have expected it, this is not a personal dig). That would have been it. Instead you have constantly mis-interpreted references, at times falsely edited them to fit your argument and tried to engage me in discussions I do not want to have and that have no relevance to the point being discussed.
My ONLY interest in this whole conversation is that you stop falsely using p-values. Other than that I do not care how you interpret the data, I am no way qualified to assess that.
The people you should be relying on for that are the coaches and exercise physiologists. Only they have the relevant experience and knowledge to interpret the data beyond the statistics and explain why it is happening. But, unfortunately, they don't agree with you.
Re: Re:
Firstly, the null hypothesis ISN'T that the groups are the same. The null hypothesis is that the difference between the groups is due to 'chance' or 'randomness', and therefore is not due to a particular 'cause' that produces an 'effect'.
If the 'methods' and 'controls' of a study are thorough enough to eliminate other possible 'causes' which might produce the observed 'effect', then a p-value which implies 'statistical significance' is typically adequate for the authors to 'claim' that the 'cause' being tested did produce the observed results.
And it seems that confusion about p-value is fairly widespread - see here for another discussion about its complexity.
http://blogs.plos.org/publichealth/2015/06/24/p-values/
Jay Kosta
Endwell NY USA
----------------------------------------------------------------------------------------------FrankDay said:
Firstly, the null hypothesis ISN'T that the groups are the same. The null hypothesis is that the difference between the groups is due to 'chance' or 'randomness', and therefore is not due to a particular 'cause' that produces an 'effect'.
If the 'methods' and 'controls' of a study are thorough enough to eliminate other possible 'causes' which might produce the observed 'effect', then a p-value which implies 'statistical significance' is typically adequate for the authors to 'claim' that the 'cause' being tested did produce the observed results.
And it seems that confusion about p-value is fairly widespread - see here for another discussion about its complexity.
http://blogs.plos.org/publichealth/2015/06/24/p-values/
Jay Kosta
Endwell NY USA
Re: Re:
They are hugely misunderstood, misinterpreted and misrepresented, although this is true of many statistical values and methods (don't get me started on PCA). A lot of the time they're not even corrected! However it's extremely well known in science and there has been a steady push to improve on these matters as big data increases the requirements for statistics in areas of science where traditionally ratios and t-tests were the norm.
So someone on a cycling forum not really getting them is completely understandable, some of the smartest people I know don't really understand them. But they actually listen when people who do understand them tell them why they can't do what they are doing...
JayKosta said:
They are hugely misunderstood, misinterpreted and misrepresented, although this is true of many statistical values and methods (don't get me started on PCA). A lot of the time they're not even corrected! However it's extremely well known in science and there has been a steady push to improve on these matters as big data increases the requirements for statistics in areas of science where traditionally ratios and t-tests were the norm.
So someone on a cycling forum not really getting them is completely understandable, some of the smartest people I know don't really understand them. But they actually listen when people who do understand them tell them why they can't do what they are doing...
Re: Re:
You say here that Powercranks were originally designed to help cyclists to better pedal in circles. You have also said that Powercranks were originally designed for HPV's. I have always claimed PC's only advantage is to train cyclists to pedal in the circular style and as circular is not as effective as mashing, PC use cannot increase power output..
FrankDay said:
You say here that Powercranks were originally designed to help cyclists to better pedal in circles. You have also said that Powercranks were originally designed for HPV's. I have always claimed PC's only advantage is to train cyclists to pedal in the circular style and as circular is not as effective as mashing, PC use cannot increase power output..
Re: Re:
The p=0.125 means that If the results were due to chance, that there is a 12.5% probability that the results that were obtained would happen. Or conversely, there's an 87.5% probability that other (less meaningful) results would happen.
What that means to the 'reader' varies ....
If the reader is looking for 'proof of validity', then 12.5% is not good enough.
If the reader is looking for an 'indication', then 12.5% might be adequate.
The p=0.125 doesn't mean that the results were due to chance, or that the items being tested did not have any effect.
Only that the 'statistics' indicated that p-value.
Jay Kosta
Endwell NY USA
----------------------------------------------------FrankDay said:
The p=0.125 means that If the results were due to chance, that there is a 12.5% probability that the results that were obtained would happen. Or conversely, there's an 87.5% probability that other (less meaningful) results would happen.
What that means to the 'reader' varies ....
If the reader is looking for 'proof of validity', then 12.5% is not good enough.
If the reader is looking for an 'indication', then 12.5% might be adequate.
The p=0.125 doesn't mean that the results were due to chance, or that the items being tested did not have any effect.
Only that the 'statistics' indicated that p-value.
Jay Kosta
Endwell NY USA
Re: Re:
Ugh, my question to you is how would you "properly" interpret the p-values given by those authors. I understand you told me my interpretation was wrong. I have asked you for what the proper interpretation of that particular data should be. The authors gave it to us, it must have some meaning.King Boonen said:
LOL. Yes, the p-value for significance is chosen before the stats are done. But, the p-value is calculated without regards to what the level chosen for "significance" is or was. If I chose 0.5 as my chosen level of significance does that suddenly turn my data into something important because 0.125 is less than that cut-off? What is one calculating when one calculates the p-value? How should it be interpreted in the study I mentioned?
Thanks again for telling me what I shouldn't do. I will await patiently for you to tell me how to properly use that data. The authors gave it to us so I suspect it isn't totally without meaning.
You are welcome to look at the entire experiment if you choose if you need to to properly interpret the data. It was my understanding that statistics looks at the data before it and calculates probabilities, etc. If I am wrong in that assessment you can correct me there also.
I wouldn't know. He managed to calculate a p<0.05 (his significance cut-off) between 145, 170 and 125, 220 and no one seemed to mind so I suspect his methods were the same between the above. The question I had was the raw data seems to show a difference between 145 and 170. Wouldn't the statistics tell us what the chance the difference seen was due to chance? Aren't you interested in finding out as if the chances are relatively low wouldn't a repeat study with more power be indicated?
Firstly, it is 120 and 220, not 120 and 195. It's right there in the graph you posted.
Secondly, I don't care, again this has nothing to do with our previous discussion and I would hazard this thread at all. How was the p-value calculated? What was compared to what? ANOVA? T-test? Were they corrected? If it's a t-test your whole question is wrong as it cannot compare two populations against two other populations. Were the populations combined? How does this relate to the initial hypothesis? Why must there be a p-value comparing 145 and 170? Was it calculated? Why are they obviously different? The study appears to only care about power from what little you presented, so they can only be assessed in such terms. If the power is not statistically different, in terms of this study, statistically, the cranks are not different either. Just like if I painted a set of cranks red. They would obviously be different to black ones, but not in terms of this experiment if power output is the same. Even if you can reject the null hypothesis, this does not mean that the alternative hypothesis is true because the statistics have no idea what the hypotheses are, only those educated in the field can decide that. The very people who are disagreeing with you (Not me. I am an analytical biochemist and I have only been pointing out your misappropriation of p-values).
You have access to the "whole experiment" so do what you need to do to answer the question. You are either lazy or using this as an excuse to answer the question because you know you won't like the answer. The author gave us the number so I expect he thought it had some meaning. You are the expert, tell us how the reader is supposed to interpret it.
Really? Do you have a link to such a definition? I found this "In inferential statistics the null hypothesis usually refers to a general statement or default position that there is no relationship between two measured phenomena, or no difference among groups.[1] Rejecting or disproving the null hypothesis—and thus concluding that there are grounds for believing that there is a relationship between two phenomena (e.g. that a potential treatment has a measurable effect)—is a central task in the modern practice of science, and gives a precise sense in which a claim is capable of being proven false." https://en.wikipedia.org/wiki/Null_hypothesis. The need for statistics is because it is essentially impossible to avoid randomness in both measuring and in the intervention aspects of any study. Calculating a p-value gives the researcher and the reader a sense of "how strong" any difference seen represents a "real" difference between the groups. A p-value of 0.04999 is hardly any more important than a p-value of 0.05001 yet one would pass the significance cut-off standard and the other wouldn't. You allow Fergie to come here and claim that some study "proves" that PowerCranks don't work because some study failed to reach the p<0.05 level even though you know that claim to be statistical hogwash. Yet, you chastise me for what you consider to be a mischaracterization of what the p-value represents but refuse to tell everyone how it should be used in these instances.
That could only occur if the two groups were identical, yet above you said that wasn't required.
It would have only lasted three posts if you had said you can't use it that way, here is the proper way to use this number. Then my third post saying thanks would have been warranted. Instead, I have been asking for that clarification for a month without any luck. So, thanks for nothing.
I can't understand how to stop using this data "falsly" unless I understand how to use it properly. That is what I have asked you to do, to educate me in this regards yet you refuse.
Let me get this straight. You believe Coach Fergie has a good handle on this statistics stuff? And, because of that I should be listening to his interpretation of these studies? LOL
Re: Re:
"They" don't listen to you because you don't tell them what they should be saying. I don't know what your academic background is but I suspect it is substantial. I would just like to point out that Dr. means teacher, at least that is what I was told in medical school. As I said, I see this way of "debating" as simply academic bullying. You point out where others are wrong (which of course makes you superior) but never say what is correct. I look forward to your eventually telling the group how one should properly interpret a p=0.125 between two groups showing a difference.King Boonen said:
One more thing. https://en.wikipedia.org/wiki/Statistical_significance
Edit: another page if you are still confused: http://www.stat.yale.edu/Courses/1997-98/101/sigtest.htm
The problem with choosing a statistical significance level goes to the statement that follows that above
Reaching the statistically significant cut-off allows the investigator, by convention, to ignore the fact that there is still a possibility that the measured difference is due to chance or if it doesn't quite reach that level there is a very good chance that there is still a measurable difference between groups. But, somehow, everyone forgets this and if a study reaches p<0.05 then it has "proved" something and if it doesn't then it "proves" the opposite. Nothing is further from the truth.
Edit: another page if you are still confused: http://www.stat.yale.edu/Courses/1997-98/101/sigtest.htm
How else is the bolded text to be interpreted other than the P-value of 0.0082 represents the probability of obtaining the results by chance? Does this calculation suddenly have no meaning if the result were 0.012 and didn't reach the P<0.01 threshold (even though above the more usual P<0.05 threshold)? When King says P doesn't represent probability what on earth does he mean? Seems to mean that according to Yale, but what would they know? He has already established I don't know anything so maybe he will come and "clarify" what they mean also.
TRENDING THREADS
-
-
Liege Bastogne Liege-7 hour monument, Sunday, April 21, 2024- Started by topcat
- Replies: 985
-
Paris - Roubaix 2024, one day monument, April 7- Started by Dreginald
- Replies: 2K
-
-
De Ronde van Vlaanderen (270.8 km), 2024 March 31 (Sunday)- Started by Netserk
- Replies: 1K
-
-
Teams & Riders The Remco Evenepoel is the next Eddy Merckx thread- Started by DNP-Old
- Replies: 24K
Latest posts
-
-
-
-
77th Tour de Romandie (2.UWT) // 23rd - 28th of April 2024- Latest: Armchair cyclist
Facebook
Twitter
Instagram