Effects can be ignorable
- oh, real, yes, but fine -
So say, worst case scenario,
μ′s right there on the line.
The power is the modest chance
Effects as small as μ′s
Will have to be significant
In the C.I. I choose.
I pick a null hypothesis
and set out to reject.
But can I shrink my sample size
and find that small effect?
I am a stingy scientist
--- Blame NIH, not me!
I pick n = 25
What will that let me see?
I set up an alternative
but have to search a bit
for standard deviations next
by trawling prior lit
I must start by considering
how many tails make sense.
Then, μ and σ both in hand,
It’s time for confidence!
Selecting my significance,
Say 95%
I find the corresponding z
For where the tails went
The sample average x̄ vs.
z times standard err’
Shows when the test rejects H-naught
Though α warns, “beware!”
A simple calculation, now
Of x̄ - μ
divided by the standard error
brings power into view
I end up with a last z score
And turn it to percent
The chance effects as small as μ
Turn out significant
Current theme: default
Less Wrong (text)
Less Wrong (link)
Arrow keys: Next/previous image
Escape or click: Hide zoomed image
Space bar: Reset image size & position
Scroll to zoom in/out
(When zoomed in, drag to pan; double-click to close)
Keys shown in yellow (e.g., ]) are accesskeys, and require a browser-specific modifier key (or keys).
]
Keys shown in grey (e.g., ?) do not require any modifier keys.
?
Esc
h
f
a
m
v
c
r
q
t
u
o
,
.
/
s
n
e
;
Enter
[
\
k
i
l
=
-
0
′
1
2
3
4
5
6
7
8
9
→
↓
←
↑
Space
x
z
`
g
Stanzas On Power Calculation
Effects can be ignorable
- oh, real, yes, but fine -
So say, worst case scenario,
μ′s right there on the line.
The power is the modest chance
Effects as small as μ′s
Will have to be significant
In the C.I. I choose.
I pick a null hypothesis
and set out to reject.
But can I shrink my sample size
and find that small effect?
I am a stingy scientist
--- Blame NIH, not me!
I pick n = 25
What will that let me see?
I set up an alternative
but have to search a bit
for standard deviations next
by trawling prior lit
I must start by considering
how many tails make sense.
Then, μ and σ both in hand,
It’s time for confidence!
Selecting my significance,
Say 95%
I find the corresponding z
For where the tails went
The sample average x̄ vs.
z times standard err’
Shows when the test rejects H-naught
Though α warns, “beware!”
A simple calculation, now
Of x̄ - μ
divided by the standard error
brings power into view
I end up with a last z score
And turn it to percent
The chance effects as small as μ
Turn out significant