Correlation/Regression

• How Far He'll Go

I've been staring at my test in statistics
Longer than I care to remember
Not been perplexed by scatterplots before
I wish I could be the perfect student
But I know it would have been prudent
If I had studied regression even more

Got a high R-squared,
What does that mean? I’m scared!
A non-zero slope
I reject H0
Cause the p is low
Errors somewhat normal I see

See the line where the jump meets the year
It confounds me
‘Cause I don’t know, how far he’ll go
If the wind on his back as he jumps stays behind him

One never knows
We must be able to figure out how far he'll go
I will use the given coefficients
Cause I know that I’m proficient
Calculating the point that’s on the line
Take the year times 0.011

Subtract 14.4 and I’m done
So maybe my answer is fine
10.3 I got
So I’ll put that dot, now my answer’s complete
But it sounds like a lot to accomplish that feat
What is wrong with me

See the line where the jump meets the year
It’s far out there
So I don’t know, how far he’ll go
So it seems extrapolation doesn’t work

For twenty-one eighty
So we don’t know
What's beyond those points
Can’t go past those points

See the line where the jump meets the year
Can’t go out there
So we don’t know, how far he’ll go
If the new point I predict leaves my plot far behind me
We don’t know
How far he'll go

• Linear Model Rap

Lyric by Mary McLellan
may sing to the tune of "Ice Ice Baby" (Vanilla Ice)

Yikes! Is it Linear, Listen…
Have you got a correlation coefficient
r is the strength of linearity
R2   is % of variability
Can we tell from r- NO! You don’t know
Residual plot- you must check
Look at residuals and see if there’s a pattern
“Statistical Evidence” only with a p-value

Linear Model…. Linear Model….

Repeat

• Losing Cause

Music and lyrics ©2017 by Lawrence Mark Lesser

I stopped buyin’ ice cream when I saw the charts:
The sales align with attacks by sharks!
And I’d go to my doctor, but won’t anymore:
His patients look sickly leavin’ his door!
Correlation information never gave me pause:
I treat it like causation – just ‘cause!

TVs per household aligns with length of life,
So I filled my rooms with that blue TV light!
Number of cops and crimes directly correlate,
So the crew in blue we should eliminate!
Correlation information confounds me and my flaws:
I treat it like causation –  just ‘cause!

Before this song, my world was causation;
But I was wrong, it was guilt by association!
So I claim my song helped -- well, maybe

• Correlation Guessing

The Correlation Guessing Game provides panels of four scatterplots and challenges you to match them with four potential values of Pearson's correlation.

• Shapesplosion

Shapesplosion is an on-line game in which a person is expected to place specifically shaped pegs into the appropriate holes within a short time period. In this project, students are asked to use the Shapesplosion game to design an experiment and collect data. This game is specifically designed so that students have the opportunity to develop and test their own unique research question. You can leave all the variables blank when you are simply trying out the game, however, if you want to find your score is the database of results, you will need to select the Participant Info box.

• TigerSAMPLING

TigerSAMPLING is almost identical to TigerSTAT. However in the TigerSAMPLING game there are additional questions that emphasize BIAS and GENERALIZABILITY. These games collect data and explore models for estimating the age of a Siberian tiger. In this game, students act as researchers on a national preserve where they are expected to catch tigers, collect data, analyze their data (using simple linear regression on transformed data), and draw appropriate conclusions

• TigerSTAT

TigerSTAT is a three dimensional on-line game where students collect data and explore models for estimating the age of a Siberian tiger. In this game, students act as researchers on a national preserve where they are expected to catch tigers, collect data, analyze their data (using simple linear regression on transformed data), and draw appropriate conclusions.