Read below: if science is a belief system and then comment on why some people ignore good science in favor of claims that confirm their own biases. What can we do to make sure we do not fall into the trap of thinking science is a belief system. What is the difference between a belief system and science?
Additionally, discuss why some people gravitate towards people who give them absolute answers as opposed to science that gives them probabilities.
Reading:
Science
Is science a belief system?
To answer this question we have to look at Scientific Reasoning.
1. Inductive reasoning
When you turn the wheel of your care you are using a type of reasoning. You reason:
If I turn this wheel, the car will turn in that direction
This is a type of Inductive reasoning. Consider another example:
The first 5 doors on this building are locked, therefore the next door will be locked.
The first statement is called a premise. The last part of the statement is the conclusion
With inductive reasoning, the conclusion does not always follow the premise. It is very possible the next door you try on the building will be open.
Inductive goes from the specific to broader generalizations
2. Deductive reasoning
This statement goes from a broad general statement (premise) to a specific statement
With deductive reasoning, if the premise is correct then the conclusion is correct.
With inductive reasoning, the conclusion does not always follow the premise.
A good logician knows if someone is making a deductive style argument they try and discredit the premise. If someone is trying to make an inductive argument, they try and discredit the conclusion
Hypo-deductive reasoning and abductive reasoning are also used but will not be tested on.
How does Science change?
Thomas Khun published a landmark book in 1962 called the structure of scientific revolutions discussing how science advances.
He claimed science was not the gradual accumulation of knowledge, but rather a sociological and thus a political phenomenon that happens in sudden paradigm shifts. The long-held ideas of the older generation were being upheld by willfully ignorant relics of the past until the current generation gains enough muster to overthrow the archaic ideas.
This was widely popularized and many adopted his ideas.
The problem with Khuns argument.
The problem with Khun’s argument was he intertwined the politics of science and the discovery of truth.
Science is inherently anti-authoritarian so while it is not political per se, if knowledge is power and science is anti-authoritarian, then those in power will see science as a threat.
The slow and steady of science.
Think of science like a massive cage death match for a wrestling federation. All the ideas get thrown in the ring and we see which ideas survive. Eventually, all other ideas die off and we tentatively accept it until another idea comes along and takes its place. Some ideas, called Theory or Law have proven over such a long period of time and have defeated so many other ideas we consider them champions.
Doing this sometimes takes a long time. Science will throw many good ideas in the ring for a long time before we find the best idea so when someone claims science does not work because science said one thing at one time and something different later, they do not understand the process of science. Science can be painstakingly slow and in a culture where we want the immediate results, science can easily get ignored and rhetoric or sales will take its place.
The trick is we need to get comfortable with uncertainty and be willing to make decisions based on the current best information.
Some people demand perfection of science and get frustrated when it does not produce an absolute truth right off the bat. This is an impossible expectation fallacy because these people are not comfortable with change or uncertainty
Additionally, science can move really fast. This usually happens when someone discovers a new truth that everyone is able to use to produce even more knowledge.
DNA is a great example. The inheritance problem was worked on for a long time but once it was solved, science burst forward solving all sorts of other problems
Pseudoscience at every corner
Because science is so powerful many people like to use it to either make money or sell ideas that are favorable to them.
The desire to claim something is science when it has not gone through the process of science is called pseudoscience.
It is important we do not confuse the two and there are several ways to help us sort it out.
The first question should always be: is it peer-reviewed?
Peer review helps minimize bias and poor science
If it is peer-reviewed, is it peer-reviewed in a quality journal
The flat earth society has made their own peer-review journals so they now have “Peer Reviewed” evidence the earth is flat. Thus we must go further than to just ask if it is peer-reviewed. We could ask:
How many times has the peer-reviewed article been cited?
Is this work from a private company or independent research?
Good science will always disclose its financing. A Pharmaceutical company may claim independent research but when the peer review is done it demonstrates the company paid for it.
Peer review constitutes primary sources
Textbooks are secondary sources
Blogs, magazines, articles, ect are considered tertiary sources
Wikipedia, Facebook, Twitter are quaternary sources
When it comes to science, we should only be using primary sources!
As an aside, Wikipedia is over 90% accurate when it comes to sciences so it is a great place to START your research but not reliable as evidence or “proof”. The two articles in Wikipedia from the sciences that are constantly being changed and MAY be the most inaccurate are those on climate change and evolution as many of the editors are not scientist
Statistics
Sample Size
To do statistics we have to assume our sample represents the population with which we want to make inferences
If we take a bunch of people who have heavily smoked marijuana their whole lives and give them a driving test after they smoke, can we get reliable evidence on the effect of marijuana on the general population?
What about the effects of marijuana on driving?
What about the effects of marijuana on heavy marijuana smokers while driving?
How we get our samples is important and can change the inferences we can make from the data.
Are the two samples different?
Let’s imagine we want to look at the length of dog hairs between two cities. We take a random sample of 1000 dogs of all different breeds and measure the length of the hair and, in one city the average hair size is 5cm and in the other city the average hair length is 6cm, are the lengths different between the cities or did the differences come about simply because of randomness in the sample? How do we know?
This introduces us to a basic T-Test or Students T-Test which is what a lot of research uses. It is basic and useable under optimal research conditions. There is a lot of research that does not fall into the optimal category but this test is what you hear about the most and this is not a statistics class so I am not going to delve into a lot of other things.
The T-test looks at the average between the two samples and uses some math to investigate and then gives us a “probability” type of answer that says there is an X percent probability the difference between the two populations (Cities) is due to random chance.
This is done with what is called a “P-Value” and generally, that value is set at 0.05% or lower before we accept there is a difference between the means. This p-value is saying there is a 5% or less probability that the difference between the two averages are not due to random chance. If you read the research, you will come across this term so it is helpful if you have a good idea about what it means
Appropriate inferences
What inferences can we make from our data or what inferences can be made from the experiment done?
This depends on a few things and even people with Ph.D.’s in statistics still get this wrong sometimes because of prior or pre-existing ideas they want to prove true. Remember, science always tries to prove itself wrong.
First, was the sample gathered randomly from the population or was it a certain group from the population? In humans, was the age group random, the ethnicity random, income levels ect. If anything was not done randomly then there is a bias that can start to creep in and the results are less reliable. Since we cannot experiment on humans we may never get perfect randomness but we try to come close and some studies do it better than others to increase the reliability of their results.
Second, was the population randomly divided between the control group and the experimental group? If not, it allows bias to creep in
So if someone is part of a group of devoutly religious people and “tested” something, they cannot make any inference to the general population so the methods on how science gathered the data are often more important than what the data says.
Using statistics for post hoc data
This is often done when we cannot directly experiment with something. The CDC does this to look at disease spread. The problem is it is not a designed experiment so we have to reign in our inferences.
Australia enacted strict Gun laws but does that mean the same thing can work over here? Maybe, maybe not, the US is not Australian so we cannot make that inference directly. One cities crime rates and gun laws cannot be inferred to other cities because it was not a random sample.
To make sense of post hoc data there must be some fancy statistics used to account for all the variables like demographics, what is the population number, how close together do they live, what is the average income, ect
Data Analysis
Science often collects a lot of data. Each piece of data is an individual fact and facts, as a level of certainty is very high, HOWEVER, it is not until we add a hypothesis, theory, or law to explain the facts that they become important.
Facts are stupid things until brought into conjunction with some higher law.
Patterns provide evidence for the process. Just because we were not present when things happen in our world does not mean we cannot know what happened with a high level of certainty.
Graphs
Graphs are summaries of data and should never be ignored. Make sure to pay attention to the axis, what are the labels, what is the range of information, what are the units?
Is there a positive correlation, is there a negative correlation
How many variables are there and what is the relationship between those variables
Graphs should simplify the data and make it clear. This is why pie charts are the worst graphs to use
So it is the METHODS of how we proceed that makes science so powerful.
This means science is something we do. It is a verb and not a noun
