where we assume,
Var ( y ij ) Total
2
2 2
We want to estimate both variance components given the data, and look at what
percentage of the variation of the total variation can be attributed to the fact that different
looms are used produce fabric in the factory.
1
Analysis in JMP
Select Analyze > Fit Model and put Strength in the Y box and Loom in the effects in
model box. The critical step is to highlight Loom in the model effects box and select
Random Effect from the Attributes pull-down menu as shown below.
E(MS) Table
E ( MS Loom ) 2 4 Loom
2
E ( MS Error ) 2
ˆ Loom
2
6.96 % variation due to looms = 78.6%
ˆ 2 1.90 % variation due to error = 21.4%
If we use the REML method we basically the same estimate along with a 95% CI for
Loom
2
.
The CI is too wide to be useful! To get precise estimates of variance components much
largersample sizes are needed.
2
Confidence Interval for Percent of Variation Due to an Effect (pgs. 489-490)
2
100(1 - )% CI for 2 is given by:
2
L 2 U
2
1 L 1U
2
where
1 MS Treatment 1 MS Treatment
L 1 and U 1 1
1
n MS E F / 2,a 1, N a
n MS
E F1 ( / 2 ), a 1, N a
Note: For unequal replicates replace n by the no whose formula is given by equation
(13-9) on pg. 487.
2
Constructing a 95% CI for for the loom data we first need the F-quantiles which
2 2
can be calculated using either the file F-quantile Calculator.JMP on the class server or
by using tables in the F-table in the appendix of your text. To find the upper F-quantile
using the table you have to use the fact that
1
F1 ( / 2 ),a 1, N a (see pg. 490 for an example)
F / 2 , N a , a 1
Thus we have,
1 29.73 1 1 29.73 1
L 1 .625 and U 1 55.633
4 1.90 4.47 4 1 .90 . 06975
which gives,
.625 2 55.633 2
2 2 and finally . 38 .98 .
1 .625 1 55.633 2 2
So looms account for between 38% and 98% of the total variation in fabric strength.
Again this interval is very wide because of the small number of replicates, however we
certainly know that loom to loom variation is not negligible.
3
Factorial Experiments with Random Effects
part
2
repeat
2
error
2
or 2
operator
2
operator
2
* part
To conduct the analysis in JMP set up the effects as you would for a two-factor factorial
design making sure to change each effect to random as shown below.
4
Output from E(MS) approach to estimating the variance components.
5
REML estimates of variance components, notice operator* part is zeroed out.
2
The Gage R & R study here produced results that indicate almost all of the variation
came from part to part variation, 92% . The fact different operators were used accounted
for less the one-tenth of a percent of the total variation and the repeatability variance
component was approximately 8%. This measurement system seems good, now we can
focus on eliminating part to part variability by using designed experiments to identify
potential sources of that variation.
Fitting the model we set up as we would for a three-factor factorial experiment and
change all effects involving B and/or C to random effects as shown below.
6
The E(MS) differ from those derived on the board because JMP uses the unrestricted
approach when handling random interaction effects. We can still use the restricted
approach, however we have to do the testing and variance component estimation “by
hand” using the Mean Squares returned by JMP.
7
Which approach is best to use? If we are willing to carefully write out the E(MS) by
hand it does not take to much extra effort analyze using the restricted model with JMP. If
you are not willing to painstakingly write out the E(MS) for your mixed model then I
would use the unrestricted approach and go with all of the results returned by JMP. Many
statistical software packages offer the use of either the restricted or unrestricted approach,
e.g. MINITAB.