I have to say your writing is becoming increasingly hilarious. I have not laughed so much since the last time I was watching a sitcom.
I love being honest with myself however, I already know & have thought a bit about correlation, causation, covariance etc. It is always nice to see an informal explanation or another deeper perspective. Please don't give up on us and each other even if we are not very self aware.
correlation: as ice cream sales increase so do rates of drowning.
Correlation DOES NOT equal causation.
Do ice cream sales cause more people to drown? No!
What could possible explain this correlated relationship? Well, maybe temperature. As it gets hotter outside, ice cream sales go up, AND - simultaneously - more people go swimming (thus more drowning rates).
I notice that I do have pre-confusion and I really have to be keeping that in my mind because I struggle with doing a question thinking that I am confident of my choice of answer but didn't realize that I actually needed to understand a specific skill first in order to tackle more of those types of questions.
so correlation is a complex phenomena (aka stated by a set of premises), that a hypothesis (a causal explanation and perhaps the conclusion of the argument) tries to explain?
I am not gonna lie to you: this lesson confused me because I had to be honest with myself. Subconsciously, we know that these things are not causes but correlations: there needs to be an explanation. For example, the concept of correlation does not equal causation was taught to me like this in middle school: There is a correlation between the increase in the crime rate and the increase in temperature. The inference here would then be, so summer is when the crime rate is the highest, and winter is when its at its lowest. There's a bunch of ways we can poke holes in this but this is how I tend to remember this concept with a silly example like this.
Whenever you might mix up correlation and causation, you can think of parallel parking as a metaphor for correlation. In parallel parking, things happen alongside each other, but that doesn't necessarily mean that one thing causes the other. It's more of an intertwined relationship where the result of one can increase, decrease, or not exist at all depending on the relationship between the two, but it doesn't suggest a cause-and-effect dynamic.
Not quite sure if this would be helpful to anyone but overall correlation does not necessarily equal causation. For example, people who tan for longer amounts of time on the beach are more likely to get attacked by a shark. While it may be the case that this is true, it may also be the case that people who tan for longer amounts of time on the beach may use scented tanning oils in order to intensify their tans. Upon making contact with the water, the scent of the tanning oil attracts sharks, increasing the likelihood of a shark attack when compared to individuals who only tan for short durations of time. The tanning oil in this case is the "confounder" or lurking variable, which explains the overall increase in likelihood for those who tan for longer to also be attacked by a shark. So, although there is an existing positive correlation between people who tan for a prolonged duration of time and shark attacks, it need not be equated with one necessarily causing the other as there is an existing confounder.
While I agree with the comments here and was also told "correlation /= causation" in my AP stats class and my college stats class, I have a hard time taking this particular lesson seriously.
I had it (probably subconsciously) at the back of my mind in the last of these lessons that "yeah, but these are just correlations", but it didn't stop from also thinking "how could I explain these as if they were a 'conclusion' regardless of whether they're actually caused by one another or not"
I think I'm simply just confused here about your concept of calling us out on "pre-confusion". Are you saying I was "pre-confused" simply because I was also practicing my ability to try and rationalize the supposed correlations being presented to me? I don't think I took them literally as explanations. What did I read wrong here? This is what I am now confused about--and that I know, that I am confused about this lesson. What here did I not understand?
I mean I went and thought "Yeah, sure, skipping can cause bad grades, and sometimes it does, but that's not exactly true. I skipped classes in high school quite frequently but still got high grades. But at the same time, years I did have bad grades didn't cause me to skip school"
When I was first learning about correlation vs causation, this example helped:
As ice cream consumption increases, so do shark attacks.
Does this mean eating ice cream causes sharks to attack you? Maybe eating ice cream gives off a particular scent that sharks enjoy, or brain freeze prevents people from swimming away from sharks... but don't start going down the rabbit hole. While the statement does not exclude the possibility of causation, it doesn't imply causation either.
In fact, it could just be correlation. Perhaps, in the summer, people eat more ice cream because it's hot. And in the summer, people swim in the ocean more...because it's hot. These two phenomena (eating ice cream and swimming in the ocean) are related because both increase due to the sweltering heat of summer. However, if it's just correlation, one doesn't necessarily cause the other.
Correlation neither points to nor excludes causation.
After being a tad nervous I'm one of those "dishonest" people JY talks about, I did my own knowledge check. Please let me know if this looks right to you!
Causation: " Walt smoked for 30 years and contracted lung cancer. Therefore, it's clear that smoking causes lung cancer"
- We observe a two seperate phenomena( walt smoked for 30 years & walt has lung cancer) and make a hypothesis about the events(smoking causes lung cancer).
Correlation: " The more cigarettes one smokes, the higher likelihood of contracting lung cancer"
- Saying that as the number of cigarettes increases, the higher the probability of contracting lung cancer. Were not saying smoking causes lung cancer but, rather observing the two events increasing together over time.
Sufficient and Necessary: " If you smoke, then you will get lung cancer"
- In this parallel universe, smoking will cause lung cancer 100% of the time. Even if you just smoke one time, it is a certainty that you will get lung cancer
A tip my chem professor taught us is to isolate the two things being compared and flip them.
So, with smoking and lung cancer rates being positively correlated, an immediate reaction would be to say smoking causes lung cancer. It's also possible that lung cancer causes one to smokes. That doesn't really make sense, and that is the point. Your brain has biases that want to make certain conclusions or connections. But it's your job as a scientist or rather as an LSAT taker to be objective and look at the evidence or explanation provided.
Wow this lesson turned personal into life advice haha. I totally agree with the ideas of honesty and of pre-confusion but have always wondered, what is a good way to check/identify if you are pre-confused? Frequent periodic testing of comprehension?
#feedback second to last paragraph was very helpful. The LSAT is a humbling experience, and if you don't own up when you are confused in the moment to catch yourself, it's going to be a hard time wondering where you steered wrong.
Let just make sure I understand this, are correlations an example of causal relationships? Or is the result of a correlation (the conclusion you get from the correlation) your causal relationship? #help
2
Topics
PT Questions
Select Preptest
You've discovered a premium feature!
Subscribe to unlock everything that 7Sage has to offer.
Hold on there, stranger! You need a free account for that.
We love that you want to get going. Just create a free account below—it only takes a minute—and then you can continue!
Hold on there, stranger! You need a free account for that.
We love that you came here to read all the amazing posts from our 300,000+ members. They all have accounts too! Just create a free account below—it only takes a minute—and then you’re free to discuss anything!
Hold on there, stranger! You need a free account for that.
We love that you want to give us feedback! Just create a free account below—it only takes a minute—and then you’re free to vote on this!
Subscribers can learn all the LSAT secrets.
Happens all the time: now that you've had a taste of the lessons, you just can't stop -- and you don't have to! Click the button.
29 comments
I have to say your writing is becoming increasingly hilarious. I have not laughed so much since the last time I was watching a sitcom.
I love being honest with myself however, I already know & have thought a bit about correlation, causation, covariance etc. It is always nice to see an informal explanation or another deeper perspective. Please don't give up on us and each other even if we are not very self aware.
correlation: as ice cream sales increase so do rates of drowning.
Correlation DOES NOT equal causation.
Do ice cream sales cause more people to drown? No!
What could possible explain this correlated relationship? Well, maybe temperature. As it gets hotter outside, ice cream sales go up, AND - simultaneously - more people go swimming (thus more drowning rates).
I notice that I do have pre-confusion and I really have to be keeping that in my mind because I struggle with doing a question thinking that I am confident of my choice of answer but didn't realize that I actually needed to understand a specific skill first in order to tackle more of those types of questions.
Should I think of correlation as not explaining how or why two things are connected? #HELP
Correlation: drinking more water increases one's visits to the bathroom
Correlation v. Causation flaw: drinking water causes one to use the bathroom more
Weakener: When drinking water, it gets completely absorbed in the body
Stregthen: Water induces one to pee
so correlation is a complex phenomena (aka stated by a set of premises), that a hypothesis (a causal explanation and perhaps the conclusion of the argument) tries to explain?
I am not gonna lie to you: this lesson confused me because I had to be honest with myself. Subconsciously, we know that these things are not causes but correlations: there needs to be an explanation. For example, the concept of correlation does not equal causation was taught to me like this in middle school: There is a correlation between the increase in the crime rate and the increase in temperature. The inference here would then be, so summer is when the crime rate is the highest, and winter is when its at its lowest. There's a bunch of ways we can poke holes in this but this is how I tend to remember this concept with a silly example like this.
Whenever you might mix up correlation and causation, you can think of parallel parking as a metaphor for correlation. In parallel parking, things happen alongside each other, but that doesn't necessarily mean that one thing causes the other. It's more of an intertwined relationship where the result of one can increase, decrease, or not exist at all depending on the relationship between the two, but it doesn't suggest a cause-and-effect dynamic.
Not quite sure if this would be helpful to anyone but overall correlation does not necessarily equal causation. For example, people who tan for longer amounts of time on the beach are more likely to get attacked by a shark. While it may be the case that this is true, it may also be the case that people who tan for longer amounts of time on the beach may use scented tanning oils in order to intensify their tans. Upon making contact with the water, the scent of the tanning oil attracts sharks, increasing the likelihood of a shark attack when compared to individuals who only tan for short durations of time. The tanning oil in this case is the "confounder" or lurking variable, which explains the overall increase in likelihood for those who tan for longer to also be attacked by a shark. So, although there is an existing positive correlation between people who tan for a prolonged duration of time and shark attacks, it need not be equated with one necessarily causing the other as there is an existing confounder.
#help
While I agree with the comments here and was also told "correlation /= causation" in my AP stats class and my college stats class, I have a hard time taking this particular lesson seriously.
I had it (probably subconsciously) at the back of my mind in the last of these lessons that "yeah, but these are just correlations", but it didn't stop from also thinking "how could I explain these as if they were a 'conclusion' regardless of whether they're actually caused by one another or not"
I think I'm simply just confused here about your concept of calling us out on "pre-confusion". Are you saying I was "pre-confused" simply because I was also practicing my ability to try and rationalize the supposed correlations being presented to me? I don't think I took them literally as explanations. What did I read wrong here? This is what I am now confused about--and that I know, that I am confused about this lesson. What here did I not understand?
I mean I went and thought "Yeah, sure, skipping can cause bad grades, and sometimes it does, but that's not exactly true. I skipped classes in high school quite frequently but still got high grades. But at the same time, years I did have bad grades didn't cause me to skip school"
When I was first learning about correlation vs causation, this example helped:
As ice cream consumption increases, so do shark attacks.
Does this mean eating ice cream causes sharks to attack you? Maybe eating ice cream gives off a particular scent that sharks enjoy, or brain freeze prevents people from swimming away from sharks... but don't start going down the rabbit hole. While the statement does not exclude the possibility of causation, it doesn't imply causation either.
In fact, it could just be correlation. Perhaps, in the summer, people eat more ice cream because it's hot. And in the summer, people swim in the ocean more...because it's hot. These two phenomena (eating ice cream and swimming in the ocean) are related because both increase due to the sweltering heat of summer. However, if it's just correlation, one doesn't necessarily cause the other.
Correlation neither points to nor excludes causation.
Incase this is helpful to anyone my poli sci courses told me to remember: correlation is not causation.
After being a tad nervous I'm one of those "dishonest" people JY talks about, I did my own knowledge check. Please let me know if this looks right to you!
Causation: " Walt smoked for 30 years and contracted lung cancer. Therefore, it's clear that smoking causes lung cancer"
- We observe a two seperate phenomena( walt smoked for 30 years & walt has lung cancer) and make a hypothesis about the events(smoking causes lung cancer).
Correlation: " The more cigarettes one smokes, the higher likelihood of contracting lung cancer"
- Saying that as the number of cigarettes increases, the higher the probability of contracting lung cancer. Were not saying smoking causes lung cancer but, rather observing the two events increasing together over time.
Sufficient and Necessary: " If you smoke, then you will get lung cancer"
- In this parallel universe, smoking will cause lung cancer 100% of the time. Even if you just smoke one time, it is a certainty that you will get lung cancer
A tip my chem professor taught us is to isolate the two things being compared and flip them.
So, with smoking and lung cancer rates being positively correlated, an immediate reaction would be to say smoking causes lung cancer. It's also possible that lung cancer causes one to smokes. That doesn't really make sense, and that is the point. Your brain has biases that want to make certain conclusions or connections. But it's your job as a scientist or rather as an LSAT taker to be objective and look at the evidence or explanation provided.
Wow this lesson turned personal into life advice haha. I totally agree with the ideas of honesty and of pre-confusion but have always wondered, what is a good way to check/identify if you are pre-confused? Frequent periodic testing of comprehension?
I got flashbacks to my Econ class in college. My professor's favorite phrase was:
"Correlation does not mean causation!"
I wasn't confused about the last lesson but I did get pre-confused reading about "pre-confused" lol.
JY with the life lessons
#feedback second to last paragraph was very helpful. The LSAT is a humbling experience, and if you don't own up when you are confused in the moment to catch yourself, it's going to be a hard time wondering where you steered wrong.
Let just make sure I understand this, are correlations an example of causal relationships? Or is the result of a correlation (the conclusion you get from the correlation) your causal relationship? #help