Philosophy final

p value or significance level

the probability of getting results at least as good as the results actually obtained (smaller values are more significant)

p value for a result to be significant

0.05 , if the result is smaller than 0.05 then it is significant

null hypothesis

the hypothesis that the test hypothesis is false

statistically significant result

some evidence in favour of being a genuine effect but not attributable to mere chance (less than 0.05)

replication

repeating a trial if a group gets a positive result in a trial

false positive

result can still be due to chance

what does knowing the false positive and false negative rate tell us

not enough information to tell us anything about a case that tests positive, u need to look at the base rate

conditional probability

standard statistical testing

tells u when a result gives some evidence in favour of a given hypothesis

bayes theorem

follows from definition of conditional probability & used to compare the impact of evidence on two hypothesis

Pr(H|E)

posterior probability of H

Pr(H)

prior probability of H

Pr(E|H)

likelihood

Pr(E)

prior probability of the evidence

the climate change campaign

fossil fuel industry ran a campaign to discredit the growing body of evidence that anthropogenic greenhouse gases were having an effect on the planet

CRU email hack (“climategate”)

emails were stolen from University of East Anglia from researchers at Climate Research Unit and made public, comments were made that suggest climate researchers are engaged in fraud and deception

stochastic terrorism

mass mediated process where hostile violence occurs repeatedly (random acts)

tobacco strategy

forming apparently independent organizations to make it appear that the message sent was the result of independent research (eg. the council for tobacco research)

refined tobacco strategy

philip morris formed TASSC to fight concerns about global warming, nuclear waste disposal, biotechnology

The advancement of Sound science coalition

sought funding from Exxon to fight concerns about global warming, nuclear waste disposal, biotechnology. employed steven milloy as director

cultural cognition

people form beliefs about climate change, not on evidence but in line with their political orientation

Old denial

global warming is not happening, human generated greenhouse gasses are not causing global warming

new denial

the impacts of global warming are beneficial or harmless, climate solutions won’t work, climate science and climate movement are unreliable

model based attribution

• computer simulations of the earths climate system

• reproduce observed warming (anthropogenic greenhouse gases r included)

• can be run again without anthropogenic greenhouse cases

ensemble

none of the models used by clmate scientists are perfectly accurate

IPCC

Intergovernmental Panel on Climate Changed

• formed by UN in 1985

• evaluate & report on the scientific literature and prepare summary reports on climate science

• NOT a research organization

• aimed at policy makers and other non specialists

IPCC statements

summaries of the scientific literature, shoe a pattern of getting increasingly stronger as the evidence for human induced climate change got stronger

claimed purpose of the tobacco industry

to find out whether smoking was dangerous and if so how to eliminate the danger from tobacco

actual practice of the tobacco industry

discrediting research on the dangers of tobacco

milankovitch cycles

eccentricity = changes in shape of earths orbit

obliquity = changes in tilt of earths rotational axis

precession = wobbles in the earths rotational axis

detection vs attribution

knowing that the climate is changing (detection)

knowing what is causing the change (attribution)

attribution studies

studies aimed at understanding the causes of some observed phenomenon

evidence of anthropogenic climate change

• nights warming faster than days

• winters warming faster than summers

• lower atmosphere warming; upper atmosphere cooling

greenhouse gases

• carbon dioxide

• methane (nonfossil and fossil)

• nitrous oxide

• nitrogen trifluoride

• sulfur hexafluoride

bayesian approach to statistics

goal = give us what we want: an assessment of how likely we should take a hypothesis to be in light of all the evidence

subjective probability

how sure you are about somehting is indicated by the odds you’d accept for a bet on it.

• degrees of belief in various hypothesis based on subjective assesment of how likely these are

• bayesian approaches estimate these by numerical subjective probabilities

bayes theoreom diagram

H is hypothesis and E is evidence

test results

target is present: positive result = true positive, negative result = false negative

target is absent: positive result = false positive, negative result = true negative

we want low false positive rate and a low false negative rate

base rate & base rate fallacy

base rate = how frequent are positive cases

• ignoring the base rate is called the base rate fallacy

reproducibility crisis

attention has been drawn to the fact that most studies that show a statistically significant effect are not reproducible

pre trial registration

online databases in which researchers can register trials before they are carried out

'p-hacking’

improper procedures (scientific misconduct)

• change # of trials after collecting data

• change hypothesis your testing

• change what your counting as an indicator of an effect

• change the pvalue for counting something statistically significant

cause for the greenhouse effect

what happens to the solar light

greenhouse effect

greenhouse gases trap some of Earth’s outgoing heat, keeping the planet warm, but too much trapping causes climate change

what happened to the global mean temperature

gone up as greenhouse gas concentrations have gone up

climate communication

a number of climate scientists engage in public outrech through books, interviews, talks, videos and social media

Merchants of doubt

number of issues:

• nuclear winter

• acid rain

• ozone depletion and regulation of cfcs

philip morris

formed the advancement of sound science coalition to fight concerns like global warming, nuclear waste disposal, biotechnology

temperature anomaly

interested in how much the climate is changing

• observed phenonmenon: increase of global temperature

Multiple models

intentionally created to have different simplifying assumptions and use different methods of modelling

• differ in their estimates of past and future climates

global climate coalition

• created in 1989 disbanded in 2002

• funded by Amoco, american forest and paper association, texaco (many more)

• 1994-2001 spent more than $63 million

• total lobbying by all environmental groups in the US: $4.7 million

global science team

• created by exxonmobil (1998)

• raises questions about who chart the future US course on global climate change

• measurement of the publics perspective on climate science is taken before the plan is launched

• victory achieved when:

  • citizens & media understand uncertainities in climate science

  • media coverage reflects balance on climate science of the validity of viewpoints that challenge the current conventional wisdom

technique of the media campaign

online harassment of climate scientists

balance as bias

press’s adherence to balance leads to biased coverage of both anthropogenic contributions to global warming and resultant action

climate of fear editorial

• climate science is sound but scientists face political attacks and mistrust

• science judged by power not evidence

• media creates confusion

• truth needs social trust to survive

anthropogenic

human impact on the environment

Publication bias

Failure to publish the results of a study on the basis of the direction or strength of the study findings

(if researchers do a study and dont get a statistically significant result they dont usually publish it)

brown and williamson memo

“doubt is our product”

• brown & williamson is a large tobacco company

• companies created doubt about scientific evidence to confuse the public and delay action instead of proving their product was safe