JMP for Recall Experiment
Set up your data table with explanatory variables; Reinforce, Isolate and Treatment.
These variables should be Data Type - Character variables. The response variable Recall
should be Data Type - Continuous. Go to the Modeling menu and choose Fit Model.
Choose Recall as the Y variable. Add Reinforce and Isolate as terms of the model. By
highlighting both Reinforce and Isolate and clicking on Cross, the interaction term will be
included in the model. Alternatively, you can highlight both Reinforce and Isolate and go
to the Macros pull down and click on Full Factorial. This will enter each main effect and
an interaction term. Then, just click on Run Model.
Response Recall
Whole Model
Summary of Fit
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observations (or Sum Wgts)
0.79292
0.735398
3.605551
24
24
1
Analysis of Variance
Source
Model
Error
C. Total
DF
5
18
23
Sum of Squares
896.0000
234.0000
1130.0000
Mean Square
179.200
13.000
F Ratio
13.7846
Prob > F
<.0001
Effect Tests
Source
Reinforce
Isolate
Reinforce*Isolate
Nparm
1
2
2
DF
1
2
2
Sum of Squares
96.00000
196.00000
604.00000
F Ratio
7.3846
7.5385
23.2308
Prob > F
0.0141
0.0042
<.0001
Residual by Predicted Plot
Recall Residual
7.5
5.0
2.5
0.0
-2.5
-5.0
-7.5
5
10
15 20 25 30
Recall Predicted
35
40
Isolate: Least Squares Means Table
Level
20
40
60
Least Sq Mean
21.500000
28.000000
22.500000
Std Error
1.2747549
1.2747549
1.2747549
Mean
21.5000
28.0000
22.5000
LSMeans Differences Student's t
Alpha = 0.050 Q = 2.10092 LSMean[i] By LSMean[j]
Mean[i]-Mean[j]
Std Err Dif
Lower CL Dif
Upper CL Dif
20
40
60
20
40
60
0
0
0
0
6.5
1.80278
2.71251
10.2875
1
1.80278
-2.7875
4.78749
-6.5
1.80278
-10.287
-2.7125
0
0
0
0
-5.5
1.80278
-9.2875
-1.7125
-1
1.80278
-4.7875
2.78749
5.5
1.80278
1.71251
9.28749
0
0
0
0
2