Sorry, you need to enable JavaScript to visit this website.

Observational Studies

  • Lyrics & Music © 2016 Monty Harper

    We've got our kids in the choir 
    The kids in the choir 
    We measured their heights
    To see how they changed
    And on the average
    Our kids got shorter 
    We took it for a fact
    Yeah, it was all arranged, but...

    CHORUS:
    No one counted on Simpson's Paradox
    Yeah no one counted on Simpson's Paradox
    Nobody counted on Simpson's Paradox
    To turn our numbers upside down

    We took the boys in the choir 
    And the girls in the choir
    We were thinking they were shorter
    Like the average had said
    But the boys got taller
    And the girls got taller
    How could this happen?
    Had we lost our heads?

    [REPEAT CHORUS] 

    Well we had more girls this year
    And fewer boys this year
    And girls are mostly shorter 
    Than boys as a rule
    So our girls got taller
    And our boys got taller
    While our choir got shorter
    And we all got schooled, because

    [REPEAT CHORUS] 

  • Lyrics and music © 2015 by Tom Toce

    It might not be that bad.
    It might not be that bad.
    Before you start to count your days,
    it might not be that bad.

    The lab results aren't vague.
    They say you've got the plague.
    But praise the skies and Thomas Bayes,
    it might not be that bad.

    Because the plague doesn't happen all that much,
    and no test's error free.
    You'd be amazed what that can mean,
    to the likes of you and me!

    It might not be that bad.
    It might not be that bad.
    Believe me, you should count the ways.
    It might not be that bad.

    The accuracy of the test may be ninety-nine percent.
    What are the odds you won't be around to pay next month's rent?

    They might not be that bad.
    It might not be that bad.
    Before you start to count your days,
    it might not be that bad!
    One more fact you need to bring: 
    The plague is a one in a million kind of thing!

    A million tests they do.
    Ten thousand say boo-hoo.
    But most are false and that is why,
    it might not be that bad.

    Because baby, it's ten thousand to one
    you haven't got the plague!

    It might not be that bad.
    It might not be that bad.

    Believe me, you should count the ways,
    it might not be that bad!
    It might not be that bad!

  • Music and Lyrics © 2016 Monty Harper

    OPENING CHORUS:
    Correlation does not imply causation
    Correlation does not imply causation

    People who own goggles are often getting fit
    We seem to see that happen quite a lot
    May we conclude that getting fit is caused by owning goggles?
    Clearly, we may not!
    If we take into account how many people swim,
    It's a factor that confounds the implication,
    For people who swim own goggles and get fit
    And that that explains our wacky correlation!

    [REPEAT OPENING CHORUS] 

    I've found that when I'm happy, I often seem to giggle
    Knowing that may really serve me well.
    Or maybe when I giggle, it tends to make me happy?
    How am I supposed to tell?
    The arrow of causation may point left or right or both,
    So here's what I suspect to be the scoop:
    I giggle and I feel happy so I giggle and feel more happy
    So I giggle because I'm in a causal loop!

    [REPEAT OPENING CHORUS] 

    The yearly rate of poverty and the length of my own hair
    Correlate with an r of .99
    I discovered this by accident and I don't know what to do
    With this amazing hair of mine!
    Is there some hidden mystic power built into my hair?
    My professor says it's nothing of the kind.
    It's coincidence my hair length predicts the yearly rate of poverty,
    The universe is just messing with my mind.

    [REPEAT OPENING CHORUS] 

    Post hoc ergo propter hoc is just a fallacy
    You expect a certain melody here, but - that's up to me 

    [REPEAT OPENING CHORUS 2X] 
  • Lyrics © Mary McLellan
    may sing to the tune of Pharell William's "Happy"
     

    It might seem crazy what I’m ‘bout to say
    There was a big fire last saturday
    Hot coals and plumes of smoke everywhere
    They sent - they sent a lot of… Firemen in there

    Lurking variables make it look like there’s a link
    Lurking variables make it look like there’s a link
    Lurking variables make it look like there’s a link
    Lurking variables make it look like there’s a link

    All over the news about this and that
    There’s too much damage and you can’t go back
    If you send a lot of firemen to help you
    Firemen will cause the damage, it’s what they do!

    Lurking variables make it look like there’s a link
    Lurking variables make it look like there’s a link
    Lurking variables make it look like there’s a link
    Lurking variables make it look like there’s a link

  • Lyrics © Mary McLellan
    may sing to the tune of "99 Bottles"

    99 variables in the experiment, 99 variables
    Confounding variables cloud the issue, you can’t claim a cause
    98 variables in the experiment, 98 variables
    Confounding variables cloud the issue, you can’t claim a cause
    You can’t claim a cause, you can’t claim a cause

  • Lyrics © Mary McLellan 
    may sing to the tune of Queen's "Another One Bites the Dust"

    Simpson’s Paradox, it does not rock
    The conclusion you get from the separate groups
    Are different when combined
    Hey! Why does it do that?
    They added those fractions wrong

    Repeat

Pages

list