We have too many researchers doing too many studies about the same topic and we incorrectly view each study as a separate event. Without quantification of the number of unpublished, published, and significant studies on a given topic, an individual study's relevance is unknown. If 200 separate researchers did 50 studies each (or 2000 did 5 studies) for a total of 10,000 studies, at.05 p, we could expect 500 false positives. When a study is published without knowing the universe of all studies on that topic, we do not know if any report of a significance level is really significant. Add that there is a bias to report and document positive over negative results. There is also data mining, where an existing database is used to search for any relationship among the historical variables at a p value and then report that relationship. With a large universe of studies and with data mining of historical data, an individual studies significance level is unknown and reproducible results is very low. Combining the data of published studies does not help, since their is a bias is what is reported and published.
The relevance of LinkedIn should be looked at from the employer's point of view. If employers search LinkedIn for candidates, and interview some, then the service is useful and job searchers should post resumes. For every LinkedIn filled job opening there will always be many more candidates that are passed over for interviews and job offers than get the job and that will make LinkedIn seem unnecessary for any specific candidate who was not interviewed based on a LinkedIn search or who gets a job through another method.
The only number that matters is total world output of carbon emissions. Countries can manipulate their carbon output to look clean through increasing imports. China produces a lot of carbon emissions but it is the largest exporter in the world. Despite the claims by many countries of switching to clean energy, many countries are also exporting their carbon output by importing high carbon output products. Germany for example, imports more than a quarter of its GDP. 10 percent of those imports come from China and 20 percent come from Asia. Germany's major imports from China are products with a high carbon output in manufacturing, such as heavy machinery, autos and auto parts. Germany also imports its electricity from other EU countries. Unless one accounts for the carbon emissions produced by imports, a focus solely on a country's measure of it own carbon emissions gives a false reading. Countries are also responsible for the carbon emissions produced by their imports. The US imports about 10 percent of its GDP. It is much more a closed economy than most other major developed countries. The US reduction in carbon emissions is a real number since its reduction is due to changes in manufacturing and energy production processes within the US and not through a switch to high carbon emission imports. Germany's switch to renewable energy has increased energy cost in Germany and made energy intensive (high carbon emission) products less competitive with imports. Importing a high carbon emission product that once was made domestically is not a true reduction in carbon emissions. The only change that occurred is the location of the carbon output.
In 1957, Vance Packard wrote The Hidden Persuaders about the use of psychological tricks to affect consumer and voter decision making and judgment. It was on NY Times Best Sellers list for a year. The New Yorker magazine's blurb about the book said, "A brisk, authoritative and frightening report on how manufacturers, fundraisers and politicians are attempting to turn the American mind into a kind of catatonic dough that will buy, give or vote at their command."
In the past, the concern was TV and radio advertising. Now it is social media. Other than a change in the way information is transmitted, what is new?
Manipulation and propaganda to influence people's decisionmaking and voting are probably as old as mankind and it will continue forever.
So much of research and news is rediscovering that many real world effects are Pareto/exponential distributions. Also known as variations of the 20-80 rule, 20 (can be 10, 30) percent of X cause 80 (can be 70, 90) percent of Y. The important point is that a few actors cause many effects. True in crime statistics, auto accidents, network internet traffic, income and wealth distributions, GDP (20 percent of countries produce 80 percent of GDP), customers and sales or profit, etc, etc. 1 percent of Reddit users and 74 percent of conflict is Pareto/ exponential equivalent to 20 percent of users cause 90 percent of conflict. If schools taught everyone statistics, and along with the binomial distribution the exponential/ Pareto distribution, we would see a lot fewer reports of the few causing many. Just as we do not see stories about the normal distribution of effects around the average.
Gary Kildall of Digital Research was well-known within the industry for his CP/M operating system at the time and CP/M was the obvious first choice for IBM PC's. Seatle Computer Products was also shipping boards with their own QDOS developed internally by Tim Paterson.
IBM would have known of both these operating systems since they were available in the computer hobbyist marketplace. It is believed IBM was negotiating for CP/M before it felt the need to seek out a second source for its operating system. It is unclear why IBM let Gates negotiate for QDOS instead of directly negotiating for the operating system.
IBM offered CP/M along with Gates' version of QDOS (MS-DOS) with its initial PC's. IBM priced CP/M at $240 and MS-DOS at $40. It appears that IBM was willing to pay per unit sales royalties to both companies. It is believed that CP/M's higher price was based on Kildall's demand for much a higher unit royalty payment than Gates demanded.
Gates became rich because IBM offered royalty payments instead of buying all rights to the software. IBM also produced a computer that could be made by third parties such as, Compaq, Dell, Gateway and many others, which allowed for increase MS-DOS sales and later increased Windows and Office sales.
Kildall did not have the privileged background that Gates did. Easier and quicker negotiations at a lower unit price by Kildall with IBM would have prevented Gates' early PC operating system success. Even without IBM's use of MS-DOS, Gates may still have achieved his wealth since it was Windows and Word (and the Office suite) which were the products that may him very wealthy and not MS-DOS.
Galileo in his day would have been deemed fake news because people believed the stars revolved around the earth.
Will Facebook remove fake news posts that have no scientific basis, such as those that say childhood vaccines cause autism, organic foods are healthier than non-organic and GMO foods are dangerous and unhealthy?
What about misleading and fake new posts comparing solar cell or wind turbine power capacity to fossil fuel and nuclear power plant capacity? Solar and wind can produce power only a few hours a day due to their need for strong sunlight and wind. They produce a third or less of their rated capacity, while fossil and nuclear fuel plants can run 24 hours a day and get very close to their stated power production capacity.
If you are an atheist, an article about a miracle is fake news. If you are religious, an article that wonderful things happen by chance is fake news.
Fake news is a mantra that really means my candidate did not get elected because unflattering things were said about my candidate by the opponent and that affected the election.
The issue is not net neutrality, but that the Obama administration had to have the Internet declared a regulated utility to give the FCC the power to impose net neutrality rules. Prior to the 1984 judicial breakup of the ATT monopoly, our long distance and local phone systems were owned and operated by ATT and its seven regional operating companies and were considered regulated public utilities. New product innovation and non-analog voice use of phone lines and connected devices were restricted and controlled by ATT, the FCC and state utility boards. Phone calls out of one's local area were very expensive. It is doubtful we would have the Internet, VOIP, etc., that we have today if ATT and government regulators still controlled the phone system. Prior to the breakup, it was illegal to link many consumer third party devices (think modem, routers, etc.) to the phone lines, unless approved by ATT and often with an expensive extra fee. In NYC in the late 1970's, when one picked up their landline phone, they were unable to complete a call because there had not been not enough available working telephone infrastructure to complete a call. Without the old ATT monopoly, we might have had some form of the public non-university Internet much sooner. Obama's government-controlled version of the Internet, under the guise of Net Neutrality, will stop Internet innovation, investment and price competition as it did previously under the old ATT publicly regulated utility concept. Let the Internet be free of public utility, government control regulation and in 10 to 20 years, today’s Internet will look as old fashion as rotary telephones and telephone copper land lines do today. Allow the Internet to continue to be an Obama government-regulated public utility under the smoke and mirrors of net neutrality and in 10-20 years it will be more expensive, unchanged, underfunded and deteriorating like most government controlled infrastructure and like the old monopolistic ATT public utility model.
Ho-hum: Another daily news story that the world is comprised of events that follow scalable exponential distributions, aka 80-20 rule, Bradford, Pareto, Zipf, etc distributions. 20% of countries produce 80% of world GDP; 20% of a company's customers produce 80% of sales; 20% of a company's products produce 80% of sales; 20% of a country's people have 80% of income, wealth; 20% of criminals account for 80% of crime; 20% of drivers have 80% of accidents, 20% of websites have 80% of internet traffic, 20% of music groups account for 80% of hits, etc, etc, etc.
it is no surprise that a few, 20%, of something produces a lot, 80%, of some output, such as a pollutant. It is the exponential nature of our world.
Because the upper tail of these common, real-life distributions are scalable, 80-20 is roughly equivalent to the top 1% accounting for 50% of whatever real life measured events are looked at. (.8x.8x.8=.51 ~.5;.2x.2.x.2 =.008 ~.01).
According to data from the World Bank (http://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC), US electric energy consumption per person (kWh per capita) was 7,517 kWh per capita in 1971 and grew by 72.6 percent to 12,973 kWh per capita in 2014. Energy efficiency programs do not decrease total energy usage. More efficient air conditioners allow more people to live and work in warmer climates. More efficient refrigerators allow more people to own bigger refrigerators, etc. Plus, additional energy savings from other appliances that we do not use more, allows us to use other energy appliances, microwaves, cable boxes, routers, rechargeable cell phones, tablets, and laptops, etc., without concern our monthly energy bill will get too high.
Diabetes and cancer have to be analyzed together. It has been known for a long time but rarely publicized that people with diabetes have higher rates of cancer and higher mortality rates from cancer. Blacks have a 75 percent higher rate of diabetes than Whites and Black women have a higher rate of diabetes than Black men.
Previous medical studies that have looked at cancer rates conditional on having diabetes have found that once having diabetes is considered, the rates of cancer mortality between Blacks and Whites is equal. Unless the researchers conditional the probability of cancer mortality on having diabetes, the unconditional data will always show higher rates of cancer deaths for Blacks. Unfortunately, researchers and media generally attribute the difference in cancer mortality between Blacks and Whites as a difference in cancer treatment and diagnosis instead of the known difference in diabetes rates, which is the primary reason for the difference in mortality.
Capacity in solar and wind energy production is misleading since neither can produce as much power per unit of capacity as fossil fuel plants can. The US EIA, https://www.eia.gov/electricit...,lists capacity factors for non-fossil fuel energy production. It also lists capacity factors for coal and fossil fuel energy production, https://www.eia.gov/electricit... .
The tables show that each unit of coal and fossil fuel energy capacity produces approximately twice as much energy over a year as the same amount of capacity of solar and wind. Until wind and solar have twice the capacity of coal, they have not overtaken coal in energy production. Even capacity factors overstate solar and wind energy production since neither wind nor solar can produce continuously power over a 24 hour day as coal and fossil fuel plants can.
Volume is a more accurate measure of how much plastic garbage is in the ocean than sq. ft. spread. The Pacific Ocean is about 64 million sq miles with an average depth of 14,000 feet. The Pacific Ocean is about 25 quintillion (25 with 18 zeros) cubic feet of water. The plastic spread is 1.35 million sq miles. Assuming a very very generous average depth of 1 ft of plastic, the plastic volume is 37.7 million cubic feet. The observed plastic garbage dump would be 1.5 millionths of the Pacific Ocean volume. If the plastic volume averaged a more realistic 1 inch depth, it would be 1.3 billionths of the Pacific Ocean volume.
US Energy Information Administration has data on monthly and yearly capacity factors (energy produced divided by potential capacity).
https://www.eia.gov/electricit...
Wind Turbines produce about 30 percent of capacity on a monthly and yearly basis. Nuclear energy produces about 90 percent of capacity.
Need about 3 times as much wind power energy generating capacity as currently have to surpass current nuclear energy generating capacity. Still have a long way to go for wind.
Research about women's poorer negotiation skills or underpayment is based on hard salary dollars and almost always does not include the value of benefits and job characteristics. For example, before the new health law, more women would prefer and accept jobs that provide better health plans at a trade-off of a lower salary, while more men would forego health benefits for higher pay. The compensation value to the employer is identical, but the reported pay of men is higher. Women as a group negotiate or choose jobs with more flex-time, less travel, less overtime, less weekend work and more on the job safety than men. This has been studied and documented by the US DOL and many researchers. Many promoting the idea of women's lower wages as compared to men fail to mention and include the value of non-W2, non-1099, job benefits and characteristics.
Whenever there is a number showing the ratio of women to men's pay, it is always an inter-employer, inter-industry number. Similar job classifications at different industries and different employers have a different ratio of benefits and job characteristic value to salary.
What has been attributed to poor salary negotiating skills of women is really women over time in the workforce, negotiating non-salary benefits or self-selecting into jobs or industries with higher amounts of non-salary benefits.
Because the large scale pay differential between men and women is always across industries or firms, just about all pay discrimination lawsuits against an individual firm employer, whether it be a Wal-Mart or a Kleiner, fails in the court.
MRIs do have false positive rates. An item may appear in an image that does not exist in the patient, is misinterpreted by the reviewing physician or does not reappear on subsequent MRI imaging. False positives create unnecessary patient anxiety, increase medical costs, result in unneeded follow up testing, unneeded medical procedures and misdiagnosis.
Increasing the speed of an MRI image must be balanced against any increase in the false positive rates.
It does not appear that any data is given on the new false positive rates and whether they remained static, increased or decreased.
In many ways, patients would gain more form a decrease in an initial false positive rate than they would from an increase in imaging speed. An increase in false positives counteracts the patient gain from an increase in speed.
Unclear what 2/3 success rate means.
Need to know false positive and false negative rates.
Does 2/3 mean of those I said were lying? In which case how many liars did I miss?
If there are 1000 liars in the sample, and I say 100 are lying, of which 66 are liars that is 2/3 correct. However, I missed 900 liars and called 34 truth tellers liars.
Or does 2/3 mean of those actually lying?
In which case how many truth tellers did I include to find the 2/3 of the liars?
If there are 1000 liars and 1000 truth tellers in a total sample of 2000 and I say 1666 are liars of which only 666 are liars that is 2/3 666/1000 liars correct and 1000 truth tellers are called liars and 334 liars are missed.
Or If I say 999 out of the 2000 are liars and only 666 are liars that is 2/3 666/999 correct with 334 liars missed and 333 truth tellers called liars..
The packaging, housing, installation, and connective materials are the major costs of a solar cell. For a long time, the cost of the photovoltaic material has not been the major economic factor inhibiting commercial use. Until someone figures out how to increase the photovoltaic output per cm2, such as in semiconductors transistors per cm2, solar cells will not achieve commercial acceptance. The new process may use less material, but a 6 inch photocell will still create about the same amount of electricity.
Lesson plans meet the definition of "work for hire" under US copyright laws and as such are owned by the school system or municipality unless there are express agreements giving the rights to the teachers. Teachers are employees and not third party contractors, such as many programmers, and lesson plans are within the scope of a teacher's employment. Lesson plans are the property of the school. State law is only relevant if it expressly gives the rights to the lesson plans to the teachers. Otherwise, the plans belong to the schools.
Survivorship bias is caused by looking at the winners at the end of a period and then tracing backwards to find causative factors.
It suffers from two failings. It does not consider those who started at the same time with the same qualities but failed to succeed. One needs to either do a forward-looking study with matched groups of identical traits and see if the groups with the traits succeed, or one has to go back and rebuild old data to include the individuals with similar traits that did not succeed. When survivorship bias is controlled for, usually statements about the cause of the effect disappear and turn out to be unsubstantiated. This is why historical medical analysis, such as taking certain vitamins to not get a disease or live longer, do not hold up when put to forward looking controlled studies. At the time of Gates' beginning, many other young people were playing with computers and programming. There were many other companies like Gates' at the same time. Some of them had the same advantages and luck as Gates', but failed.
The second failing is data mining. It occurs when one uses historical data and attempts to find some causative relationships. For example, if Gladwell looked at 100 successful individuals, the average odds are, let us say at a ten percent statistical confidence level, that 10 people will fit his criteria. It is these ten people he writes about, but it is nothing but the effect of probability. Of course, he does not mention the ones he looked at that he could not fit into his preconceived idea. It is like taking 100 coins and separately recording the results of flipping each one. About 12 coins, on average, will come up with three heads each in the first three tosses. To later say that these coins are different or more successful than the others without more info is meaningless. This is a common problem with stock market advice. Looking back over any period one can find stock relationships. However, those relationships will no longer hold going forward and were just a statistical fluke of the old data. This is why people who think they have found a pattern in the stock market can never consistently make money later.
Slashdot Mirror
We have too many researchers doing too many studies about the same topic and we incorrectly view each study as a separate event. Without quantification of the number of unpublished, published, and significant studies on a given topic, an individual study's relevance is unknown. If 200 separate researchers did 50 studies each (or 2000 did 5 studies) for a total of 10,000 studies, at .05 p, we could expect 500 false positives. When a study is published without knowing the universe of all studies on that topic, we do not know if any report of a significance level is really significant. Add that there is a bias to report and document positive over negative results. There is also data mining, where an existing database is used to search for any relationship among the historical variables at a p value and then report that relationship. With a large universe of studies and with data mining of historical data, an individual studies significance level is unknown and reproducible results is very low. Combining the data of published studies does not help, since their is a bias is what is reported and published.
The relevance of LinkedIn should be looked at from the employer's point of view. If employers search LinkedIn for candidates, and interview some, then the service is useful and job searchers should post resumes. For every LinkedIn filled job opening there will always be many more candidates that are passed over for interviews and job offers than get the job and that will make LinkedIn seem unnecessary for any specific candidate who was not interviewed based on a LinkedIn search or who gets a job through another method.
The only number that matters is total world output of carbon emissions. Countries can manipulate their carbon output to look clean through increasing imports. China produces a lot of carbon emissions but it is the largest exporter in the world. Despite the claims by many countries of switching to clean energy, many countries are also exporting their carbon output by importing high carbon output products. Germany for example, imports more than a quarter of its GDP. 10 percent of those imports come from China and 20 percent come from Asia. Germany's major imports from China are products with a high carbon output in manufacturing, such as heavy machinery, autos and auto parts. Germany also imports its electricity from other EU countries. Unless one accounts for the carbon emissions produced by imports, a focus solely on a country's measure of it own carbon emissions gives a false reading. Countries are also responsible for the carbon emissions produced by their imports. The US imports about 10 percent of its GDP. It is much more a closed economy than most other major developed countries. The US reduction in carbon emissions is a real number since its reduction is due to changes in manufacturing and energy production processes within the US and not through a switch to high carbon emission imports. Germany's switch to renewable energy has increased energy cost in Germany and made energy intensive (high carbon emission) products less competitive with imports. Importing a high carbon emission product that once was made domestically is not a true reduction in carbon emissions. The only change that occurred is the location of the carbon output.
In 1957, Vance Packard wrote The Hidden Persuaders about the use of psychological tricks to affect consumer and voter decision making and judgment. It was on NY Times Best Sellers list for a year. The New Yorker magazine's blurb about the book said, "A brisk, authoritative and frightening report on how manufacturers, fundraisers and politicians are attempting to turn the American mind into a kind of catatonic dough that will buy, give or vote at their command." In the past, the concern was TV and radio advertising. Now it is social media. Other than a change in the way information is transmitted, what is new? Manipulation and propaganda to influence people's decisionmaking and voting are probably as old as mankind and it will continue forever.
So much of research and news is rediscovering that many real world effects are Pareto/exponential distributions. Also known as variations of the 20-80 rule, 20 (can be 10, 30) percent of X cause 80 (can be 70, 90) percent of Y. The important point is that a few actors cause many effects. True in crime statistics, auto accidents, network internet traffic, income and wealth distributions, GDP (20 percent of countries produce 80 percent of GDP), customers and sales or profit, etc, etc. 1 percent of Reddit users and 74 percent of conflict is Pareto/ exponential equivalent to 20 percent of users cause 90 percent of conflict. If schools taught everyone statistics, and along with the binomial distribution the exponential/ Pareto distribution, we would see a lot fewer reports of the few causing many. Just as we do not see stories about the normal distribution of effects around the average.
Gary Kildall of Digital Research was well-known within the industry for his CP/M operating system at the time and CP/M was the obvious first choice for IBM PC's. Seatle Computer Products was also shipping boards with their own QDOS developed internally by Tim Paterson. IBM would have known of both these operating systems since they were available in the computer hobbyist marketplace. It is believed IBM was negotiating for CP/M before it felt the need to seek out a second source for its operating system. It is unclear why IBM let Gates negotiate for QDOS instead of directly negotiating for the operating system. IBM offered CP/M along with Gates' version of QDOS (MS-DOS) with its initial PC's. IBM priced CP/M at $240 and MS-DOS at $40. It appears that IBM was willing to pay per unit sales royalties to both companies. It is believed that CP/M's higher price was based on Kildall's demand for much a higher unit royalty payment than Gates demanded. Gates became rich because IBM offered royalty payments instead of buying all rights to the software. IBM also produced a computer that could be made by third parties such as, Compaq, Dell, Gateway and many others, which allowed for increase MS-DOS sales and later increased Windows and Office sales. Kildall did not have the privileged background that Gates did. Easier and quicker negotiations at a lower unit price by Kildall with IBM would have prevented Gates' early PC operating system success. Even without IBM's use of MS-DOS, Gates may still have achieved his wealth since it was Windows and Word (and the Office suite) which were the products that may him very wealthy and not MS-DOS.
Galileo in his day would have been deemed fake news because people believed the stars revolved around the earth. Will Facebook remove fake news posts that have no scientific basis, such as those that say childhood vaccines cause autism, organic foods are healthier than non-organic and GMO foods are dangerous and unhealthy? What about misleading and fake new posts comparing solar cell or wind turbine power capacity to fossil fuel and nuclear power plant capacity? Solar and wind can produce power only a few hours a day due to their need for strong sunlight and wind. They produce a third or less of their rated capacity, while fossil and nuclear fuel plants can run 24 hours a day and get very close to their stated power production capacity. If you are an atheist, an article about a miracle is fake news. If you are religious, an article that wonderful things happen by chance is fake news. Fake news is a mantra that really means my candidate did not get elected because unflattering things were said about my candidate by the opponent and that affected the election.
The US Supreme Court should broadcast and stream all its hearings live.
The issue is not net neutrality, but that the Obama administration had to have the Internet declared a regulated utility to give the FCC the power to impose net neutrality rules. Prior to the 1984 judicial breakup of the ATT monopoly, our long distance and local phone systems were owned and operated by ATT and its seven regional operating companies and were considered regulated public utilities. New product innovation and non-analog voice use of phone lines and connected devices were restricted and controlled by ATT, the FCC and state utility boards. Phone calls out of one's local area were very expensive. It is doubtful we would have the Internet, VOIP, etc., that we have today if ATT and government regulators still controlled the phone system. Prior to the breakup, it was illegal to link many consumer third party devices (think modem, routers, etc.) to the phone lines, unless approved by ATT and often with an expensive extra fee. In NYC in the late 1970's, when one picked up their landline phone, they were unable to complete a call because there had not been not enough available working telephone infrastructure to complete a call. Without the old ATT monopoly, we might have had some form of the public non-university Internet much sooner. Obama's government-controlled version of the Internet, under the guise of Net Neutrality, will stop Internet innovation, investment and price competition as it did previously under the old ATT publicly regulated utility concept. Let the Internet be free of public utility, government control regulation and in 10 to 20 years, today’s Internet will look as old fashion as rotary telephones and telephone copper land lines do today. Allow the Internet to continue to be an Obama government-regulated public utility under the smoke and mirrors of net neutrality and in 10-20 years it will be more expensive, unchanged, underfunded and deteriorating like most government controlled infrastructure and like the old monopolistic ATT public utility model.
Ho-hum: Another daily news story that the world is comprised of events that follow scalable exponential distributions, aka 80-20 rule, Bradford, Pareto, Zipf, etc distributions. 20% of countries produce 80% of world GDP; 20% of a company's customers produce 80% of sales; 20% of a company's products produce 80% of sales; 20% of a country's people have 80% of income, wealth; 20% of criminals account for 80% of crime; 20% of drivers have 80% of accidents, 20% of websites have 80% of internet traffic, 20% of music groups account for 80% of hits, etc, etc, etc. it is no surprise that a few, 20%, of something produces a lot, 80%, of some output, such as a pollutant. It is the exponential nature of our world. Because the upper tail of these common, real-life distributions are scalable, 80-20 is roughly equivalent to the top 1% accounting for 50% of whatever real life measured events are looked at. (.8x.8x.8= .51 ~ .5; .2x.2.x.2 = .008 ~ .01).
According to data from the World Bank (http://data.worldbank.org/indicator/EG.USE.ELEC.KH.PC), US electric energy consumption per person (kWh per capita) was 7,517 kWh per capita in 1971 and grew by 72.6 percent to 12,973 kWh per capita in 2014. Energy efficiency programs do not decrease total energy usage. More efficient air conditioners allow more people to live and work in warmer climates. More efficient refrigerators allow more people to own bigger refrigerators, etc. Plus, additional energy savings from other appliances that we do not use more, allows us to use other energy appliances, microwaves, cable boxes, routers, rechargeable cell phones, tablets, and laptops, etc., without concern our monthly energy bill will get too high.
Diabetes and cancer have to be analyzed together. It has been known for a long time but rarely publicized that people with diabetes have higher rates of cancer and higher mortality rates from cancer. Blacks have a 75 percent higher rate of diabetes than Whites and Black women have a higher rate of diabetes than Black men. Previous medical studies that have looked at cancer rates conditional on having diabetes have found that once having diabetes is considered, the rates of cancer mortality between Blacks and Whites is equal. Unless the researchers conditional the probability of cancer mortality on having diabetes, the unconditional data will always show higher rates of cancer deaths for Blacks. Unfortunately, researchers and media generally attribute the difference in cancer mortality between Blacks and Whites as a difference in cancer treatment and diagnosis instead of the known difference in diabetes rates, which is the primary reason for the difference in mortality.
Capacity in solar and wind energy production is misleading since neither can produce as much power per unit of capacity as fossil fuel plants can. The US EIA, https://www.eia.gov/electricit... ,lists capacity factors for non-fossil fuel energy production. It also lists capacity factors for coal and fossil fuel energy production, https://www.eia.gov/electricit... .
The tables show that each unit of coal and fossil fuel energy capacity produces approximately twice as much energy over a year as the same amount of capacity of solar and wind. Until wind and solar have twice the capacity of coal, they have not overtaken coal in energy production. Even capacity factors overstate solar and wind energy production since neither wind nor solar can produce continuously power over a 24 hour day as coal and fossil fuel plants can.
Volume is a more accurate measure of how much plastic garbage is in the ocean than sq. ft. spread. The Pacific Ocean is about 64 million sq miles with an average depth of 14,000 feet. The Pacific Ocean is about 25 quintillion (25 with 18 zeros) cubic feet of water. The plastic spread is 1.35 million sq miles. Assuming a very very generous average depth of 1 ft of plastic, the plastic volume is 37.7 million cubic feet. The observed plastic garbage dump would be 1.5 millionths of the Pacific Ocean volume. If the plastic volume averaged a more realistic 1 inch depth, it would be 1.3 billionths of the Pacific Ocean volume.
US Energy Information Administration has data on monthly and yearly capacity factors (energy produced divided by potential capacity). https://www.eia.gov/electricit... Wind Turbines produce about 30 percent of capacity on a monthly and yearly basis. Nuclear energy produces about 90 percent of capacity. Need about 3 times as much wind power energy generating capacity as currently have to surpass current nuclear energy generating capacity. Still have a long way to go for wind.
Research about women's poorer negotiation skills or underpayment is based on hard salary dollars and almost always does not include the value of benefits and job characteristics. For example, before the new health law, more women would prefer and accept jobs that provide better health plans at a trade-off of a lower salary, while more men would forego health benefits for higher pay. The compensation value to the employer is identical, but the reported pay of men is higher. Women as a group negotiate or choose jobs with more flex-time, less travel, less overtime, less weekend work and more on the job safety than men. This has been studied and documented by the US DOL and many researchers. Many promoting the idea of women's lower wages as compared to men fail to mention and include the value of non-W2, non-1099, job benefits and characteristics. Whenever there is a number showing the ratio of women to men's pay, it is always an inter-employer, inter-industry number. Similar job classifications at different industries and different employers have a different ratio of benefits and job characteristic value to salary. What has been attributed to poor salary negotiating skills of women is really women over time in the workforce, negotiating non-salary benefits or self-selecting into jobs or industries with higher amounts of non-salary benefits. Because the large scale pay differential between men and women is always across industries or firms, just about all pay discrimination lawsuits against an individual firm employer, whether it be a Wal-Mart or a Kleiner, fails in the court.
MRIs do have false positive rates. An item may appear in an image that does not exist in the patient, is misinterpreted by the reviewing physician or does not reappear on subsequent MRI imaging. False positives create unnecessary patient anxiety, increase medical costs, result in unneeded follow up testing, unneeded medical procedures and misdiagnosis. Increasing the speed of an MRI image must be balanced against any increase in the false positive rates. It does not appear that any data is given on the new false positive rates and whether they remained static, increased or decreased. In many ways, patients would gain more form a decrease in an initial false positive rate than they would from an increase in imaging speed. An increase in false positives counteracts the patient gain from an increase in speed.
Unclear what 2/3 success rate means. Need to know false positive and false negative rates. Does 2/3 mean of those I said were lying? In which case how many liars did I miss? If there are 1000 liars in the sample, and I say 100 are lying, of which 66 are liars that is 2/3 correct. However, I missed 900 liars and called 34 truth tellers liars. Or does 2/3 mean of those actually lying? In which case how many truth tellers did I include to find the 2/3 of the liars? If there are 1000 liars and 1000 truth tellers in a total sample of 2000 and I say 1666 are liars of which only 666 are liars that is 2/3 666/1000 liars correct and 1000 truth tellers are called liars and 334 liars are missed. Or If I say 999 out of the 2000 are liars and only 666 are liars that is 2/3 666/999 correct with 334 liars missed and 333 truth tellers called liars..
The packaging, housing, installation, and connective materials are the major costs of a solar cell. For a long time, the cost of the photovoltaic material has not been the major economic factor inhibiting commercial use. Until someone figures out how to increase the photovoltaic output per cm2, such as in semiconductors transistors per cm2, solar cells will not achieve commercial acceptance. The new process may use less material, but a 6 inch photocell will still create about the same amount of electricity.
Lesson plans meet the definition of "work for hire" under US copyright laws and as such are owned by the school system or municipality unless there are express agreements giving the rights to the teachers. Teachers are employees and not third party contractors, such as many programmers, and lesson plans are within the scope of a teacher's employment. Lesson plans are the property of the school. State law is only relevant if it expressly gives the rights to the lesson plans to the teachers. Otherwise, the plans belong to the schools.
Survivorship bias is caused by looking at the winners at the end of a period and then tracing backwards to find causative factors. It suffers from two failings. It does not consider those who started at the same time with the same qualities but failed to succeed. One needs to either do a forward-looking study with matched groups of identical traits and see if the groups with the traits succeed, or one has to go back and rebuild old data to include the individuals with similar traits that did not succeed. When survivorship bias is controlled for, usually statements about the cause of the effect disappear and turn out to be unsubstantiated. This is why historical medical analysis, such as taking certain vitamins to not get a disease or live longer, do not hold up when put to forward looking controlled studies. At the time of Gates' beginning, many other young people were playing with computers and programming. There were many other companies like Gates' at the same time. Some of them had the same advantages and luck as Gates', but failed.
The second failing is data mining. It occurs when one uses historical data and attempts to find some causative relationships. For example, if Gladwell looked at 100 successful individuals, the average odds are, let us say at a ten percent statistical confidence level, that 10 people will fit his criteria. It is these ten people he writes about, but it is nothing but the effect of probability. Of course, he does not mention the ones he looked at that he could not fit into his preconceived idea. It is like taking 100 coins and separately recording the results of flipping each one. About 12 coins, on average, will come up with three heads each in the first three tosses. To later say that these coins are different or more successful than the others without more info is meaningless. This is a common problem with stock market advice. Looking back over any period one can find stock relationships. However, those relationships will no longer hold going forward and were just a statistical fluke of the old data. This is why people who think they have found a pattern in the stock market can never consistently make money later.