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

# Experiments

• ### Gender Neutral Design

Q. What's a row-column design for two blocking factors that does not distinguish gender?

A. A Latinx Square!

Larry Lesser

• ### Latin American Square

A B C D E
B C E A D
C E D B A
D A B E C
E D A C B

Where A = Argentina; B = Brazil; C = Colombia; D = Dominican Republic; and E = El Salvador

Larry Lesser

• ### Randomly Select

may sing to the tune of Meghan Trainor's "All About That Bass"

You know you randomly select, select, no effect
You know you randomly select, select, no effect
You know you randomly select, select, no effect
When your sample’s like the population you can generalize

You know you randomly select, select, no effect
You know you randomly select, select, no effect
You know you randomly select, select, no effect
When your sample’s like the population you can generalize

When you make an experiment you must randomly assign
To make the groups the same so your treatment is the why
When you make an experiment you must randomly assign
To make the groups the same so your treatment is the why

• ### Randomize

may sing to the tune of The Bee Gee's "Stayin' Alive"

Well you can tell by the way I design my experiment
I use my mind
Treatment A and Treatment B of the experiment is what you’ll see
Because subjects-- they go in my big bag I shake them well
Subjects-- they go in my big bag I shake them well

When you need to spread unknown traits evenly - Randomize, RANDOMIZE
When you need to spread unknown traits evenly - Randomize, RANDOMIZE
RANDOMIZE to REDUCE BIAS, REDUCE BIAS
RANDOMIZE to REDUCE BIAS

• ### Principles of Experimental Design

may sing to the tune of "Spongebob Squarepants Theme Song"

Make an experiment what do you do?
Control, randomize, repeat

Make an experiment what do you do?
Control, randomize, repeat

Make an experiment what do you do?
Control, randomize, repeat

Make an experiment what do you do?
Control, randomize, repeat

Randomly assign to the treatment groups
So you can claim the cause and effect

• ### Blocking

may sing to the tune of The Beach Boys "Good Vibrations"

I see you have a trait there
That may affect the way subjects respond
I see you have a trait there
That may affect the way subjects respond
What do you do
Make homogeneous groups
What do you do
Make homogeneous groups

Blocking reduces variation
Blocking reduces variations
Within each block is experimentation
Within each block is experimentation