The Rural Technology Initiative ceased operations in 2011. This site is maintained as an archive of works from RTI collaborators from 2000 to 2011 and is no longer updated. RTI's successor and remaining staff can be found at NRSIG.org


     
 
   
Search the RTI Website
 
Click to go to the Precision Forestry Cooperative website
Click to go to the RTI Home page
Click to go to the About RTI page
Click to go to the RTI Projects page
Click to go to the RTI Publications page
Click to go to the RTI Tools page
Click to go to the RTI Geographic Information Systems page
Click to go to the RTI Streaming Video Directory
Click to go to the RTI Training page
Click to go to the RTI Contacts page
Click to go to the RTI Image Archive
Click to go to the RTI Site Map
Click to go to the RTI Links page


Appendix I

Statistics For Binary Linear Regression

The statistics were broken into steps to determine which variables would best work for an equation. All statistics were done in SPSS. Below are the significant steps taken and the variables in each step. Step 3 is the most significant.

Step 1: Log10(Basize)
Step 2: Log10(Basize), Precipitation
Step 3: Log10(Basize), Precipitation, Slope

Elevation, downstream gradient and site class are not part of the steps due to it being insignificant for the model.

Hosmer and Lemeshow Test
Step
Chi-square
df
Sig.
1
8.864
7
0.263
2
9.9
8
0.272
3
10.262
8
0.247
Appendix Table I1. The Hosmer-Lemeshow statistic. Indicates a poor fit if the significance value is less than 0.05.


Contingency Table for Hosmer and Lemeshow Test
Head = .00
Head = 1.00
 
   
Observed
Expected
Observed
Expected
Total
Step 1 1
14
13.289
0
0.711
14
  2
10
9.966
1
1.034
11
  3
8
9.082
3
1.918
11
  4
8
7.142
3
3.858
11
  5
3
5.195
8
5.805
11
  6
4
3.662
7
7.338
11
  7
2
2.283
9
8.717
11
  8
4
1.572
7
9.428
11
  9
0
0.81
15
14.19
15
             
Step 2 1
11
10.632
0
0.368
11
  2
11
10.225
0
0.775
11
  3
10
9.602
1
1.398
11
  4
6
7.908
5
3.092
11
  5
6
5.435
5
5.565
11
  6
2
4.162
9
6.838
11
  7
2
2.679
9
8.321
11
  8
4
1.567
7
9.433
11
  9
1
0.644
10
10.356
11
  10
0
0.147
7
6.853
7
             
Step 3 1
11
10.903
0
0.097
11
  2
11
10.543
0
0.457
11
  3
11
9.741
0
1.259
11
  4
6
7.948
5
3.052
11
  5
4
5.371
7
5.629
11
  6
4
3.887
7
7.113
11
  7
2
2.508
9
8.492
11
  8
4
1.466
7
9.534
11
  9
0
0.554
11
10.446
11
  10
0
0.079
7
6.921
7
Appendix Table I2. This statistic is the most reliable test of model fit for SPSS binary logistic regression, because it aggregates the observations into groups of cases.

 

 

Appendix Figure I1

Appendix Figure I1. Deviance plot change helps identify cases that are poorly fit by the model.

 

 

Appendix Figure I2

Appendix Figure I2. The shape of the Cook's distances plot generally follows that of the previous figure, with some minor exceptions. These exceptions are high-leverage points, and can be influential to the analysis.

 

Variables not in the Equation
Score
df
Sig.
Step 1 Variables PRECIP
4.452
1
0.035
    Slope
3.242
1
0.072
    Elevation
2.268
1
0.132
    Downgrad
0.611
1
0.434
    Site Class
0.152
1
0.696
  Overall Statistics  
11.696
5
0.039
Step 2 Variables Slope
6.284
1
0.012
    Elevation
0.45
1
0.502
    Downgrad
2.153
1
0.142
    Site Class
0.148
1
0.7
  Overall Statistics  
7.889
4
0.096
Step 3 Variables Elevation
0.644
1
0.422
    Downgrad
1.126
1
0.289
    Site Class
0.121
1
0.728
  Overall Statistics  
1.658
3
0.646
Appendix Table I3. Forward stepwise methods variables left from steps

 

Model if Term Removed
Variable Model Log
Change in -2 Log
 
Sig. of the
    Likelihood
Likelihood
df
Change
Step 1 Log10(BASIZE) -73.474
55.107
1
0
Step 2 Log10(BASIZE) -73.455
59.991
1
0
  PRECIP -45.92
4.921
1
0.027
Step 3 Log10(BASIZE) -73.422
66.715
1
0
  PRECIP -44.246
8.363
1
0.004
  Slope -43.46
6.789
1
0.009
Appendix Table I4. The variables chosen by the forward stepwise method all having significant changes in -2 log-likelihood.

 

Appendix Table I5. The pseudo r-squared statistics.

 

Classification Table(a)
     
Predicted
     
Head
Percentage
Observed
0
1
Correct
Step 1 Head
0
41
12
77.358
   
1
8
45
84.906
  Overall Percentage      
81.132
Step 2 Head
0
40
13
75.472
   
1
9
44
83.019
  Overall Percentage      
79.245
Step 3 Head
0
41
12
77.358
   
1
6
47
88.679
  Overall Percentage      
83.019
Appendix Table I6. The classification table indicating the practical results of using the logistic regression model.

 

Variables in the Equation
           
95.0% C.I.for EXP(B)
   
B
S.E.
Wald
df
Sig.
Exp(B)
Lower
Upper
Step 1(a) Log10(BASIZE)
5.115
0.935
29.948
1
0
166.464
26.653
1039.659
  Constant
-2.711
0.57
22.617
1
0
0.066
   
Step 2(b) Log10(BASIZE)
5.753
1.064
29.241
1
0
315.028
39.157
2534.451
  PRECIP
0.336
0.165
4.157
1
0.041
1.4
1.013
1.934
  Constant
-30.754
13.849
4.931
1
0.026
0
   
Step 3(c) Log10(BASIZE)
7.235
1.425
25.766
1
0
1386.737
84.879
22656.314
  PRECIP
0.477
0.184
6.697
1
0.01
1.612
1.123
2.313
  SLOPE
0.096
0.04
5.786
1
0.016
1.101
1.018
1.191
  Constant
-45.172
16.05
7.921
1
0.005
0
   
Appendix Table I7. The parameter estimates table summarizing the effect of each predictor.

 

 

Appendix Figure I3

Appendix Figure I3. Boxplots comparing the distribution of % slope and basin size values for PIP’s.

 

 
School of Environmental and Forest Sciences
USDA Forest Service State & Private Forestry
WSU Cooperative Extension
The Rural Technology Home Page is provided by the College of Forest Resources. For more information, please contact the Rural Technology Initiative, University of Washington Box 352100 Seattle, WA 98195, (206) 543-0827. © 2000-2004, University of Washington, Rural Technology Initiative, including all photographs and images unless otherwise noted. To view the www.ruraltech.org privacy policy, click here.
Last Updated 10/13/2022 12:34:39 PM