ANA BİLEŞENLERİN ÖZ VE VARYANS DEĞERLERİ
Eigenvalues
Function Eigenvalue % of Variance Cumulative % Canonical Correlation
1 1,589 73,5 73,5 0,783
2 0,572 26,5 100 0,603
a  First 2 canonical discriminant functions were used in the analysis.
Classification Results
      Predicted Group Membership     Total
    BOLGE malatya ege kayseri  
Original Count malatya 24 5 1 30
    ege 4 22 4 30
    kayseri 1 4 25 30
  % malatya 80 16,7 3,3 100
    ege 13,3 73,3 13,3 100
    kayseri 3,3 13,3 83,3 100
Cross-validated Count malatya 22 6 2 30
    ege 7 18 5 30
    kayseri 1 7 22 30
  % malatya 73,3 20 6,7 100
    ege 23,3 60 16,7 100
    kayseri 3,3 23,3 73,3 100
a  Cross validation is done only for those cases in the analysis. In cross validation, each case is classified by the functions derived from all cases other than that case.
b  78,9% of original grouped cases correctly classified.
c  68,9% of cross-validated grouped cases correctly classified.
ANOVA
Multiple Comparisons
Dependent Variable: TOPLAM 
Tukey HSD 
    Mean Difference (I-J) Std. Error Sig. 95% Confidence Interval  
(I) BOLGE (J) BOLGE       Lower Bound Upper Bound
malatya ege -63,2117 14,51766 0 -97,8288 -28,5947
  kayseri 6,3473 14,51766 0,9 -28,2697 40,9644
ege malatya 63,2117 14,51766 0 28,5947 97,8288
  kayseri 69,5591 14,51766 0 34,942 104,1761
kayseri malatya -6,3473 14,51766 0,9 -40,9644 28,2697
  ege -69,5591 14,51766 0 -104,1761 -34,942
*  The mean difference is significant at the .05 level.
STEPWISE DİSKRİMİNANT ANALİZİ (F 3,84 ENTRY 2,71 REMOVAL)
Variables Entered/Removed
Entered Wilks' Lambda
    Statistic df1 df2 df3 Exact F    
Step           Statistic df1 df2
1 FLEG1 0,718598562 1 2 87 17,034494 2 87
2 WINGLENG 0,485050493 2 2 87 18,741188 4 172
3 HINDWING 0,43854863 3 2 87 14,451396 6 170
At each step, the variable that minimizes the overall Wilks' Lambda is entered.
a Maximum number of steps is 34.
b Minimum partial F to enter is 3.84.
c Maximum partial F to remove is 2.71.
d F level, tolerance, or VIN insufficient for further computation.
Variables in the Analysis
Step   Tolerance F to Remove Wilks' Lambda
1 FLEG1 1 17,03449354  
2 FLEG1 0,678996913 28,13967254 0,8024729
  WINGLENG 0,678996913 20,70416819 0,7185986
3 FLEG1 0,678737369 27,15263661 0,718731
  WINGLENG 0,308154195 9,173472424 0,5332078
  HINDWING 0,365346038 4,506522263 0,4850505
Wilks' Lambda
  Number of Variables Lambda df1 df2 df3 Exact F      
Step           Statistic df1 df2 Sig.
1 1 0,718598562 1 2 87 17,034494 2 87 5,72E-07
2 2 0,485050493 2 2 87 18,741188 4 172 8,35E-13
3 3 0,43854863 3 2 87 14,451396 6 170 2,73E-13
Eigenvalues
Function Eigenvalue % of Variance Cumulative % Canonical Correlation
1 0,746171262 70,9269656 70,9269656 0,653696
2 0,305856349 29,0730344 100 0,4839618
a First 2 canonical discriminant functions were used in the analysis.
Wilks' Lambda
Test of Function(s) Wilks' Lambda Chi-square df Sig.
1 through 2 0,43854863 70,88847326 6 2,687E-13
2 0,765781014 22,94987675 2 1,039E-05
Classification Function Coefficients
BOLGE
  malatya ege kayseri
WINGLENG 1,025394688 0,959509928 0,840592396
HINDWING 1,048860913 1,214857938 1,15162161
FLEG1 0,085405741 0,264008511 0,530438079
(Constant) -230,476112 -253,8143657 -229,4721253
Fisher's linear discriminant functions
Classification Results
      Predicted Group Membership     Total
    BOLGE malatya ege kayseri  
Original Count malatya 23 4 3 30
    ege 6 20 4 30
    kayseri 1 4 25 30
  % malatya 76,66666667 13,333333 10 100
    ege 20 66,666667 13,3333333 100
    kayseri 3,333333333 13,333333 83,3333333 100
Cross-validated Count malatya 22 4 4 30
    ege 6 19 5 30
    kayseri 2 4 24 30
  % malatya 73,33333333 13,333333 13,3333333 100
    ege 20 63,333333 16,6666667 100
    kayseri 6,666666667 13,333333 80 100
a Cross validation is done only for those cases in the analysis. In cross validation, each case is classified by the functions derived from all cases other than that case.
b 75,6% of original grouped cases correctly classified.
c 72,2% of cross-validated grouped cases correctly classified.
STEPWISE DİSKRİMİNANT ANALİZİ (F 0,05 ENTRY 0,1 REMOVAL)
Variables Entered/Removed
  Entered Removed Wilks' Lambda              
      Statistic df1 df2 df3 Exact F      
Step             Statistic df1 df2 Sig.
1 FLEG1   0,718598562 1 2 87 17,03449 2 87 5,72E-07
2 WINGLENG   0,485050493 2 2 87 18,74119 4 172 8,35E-13
3 HINDWING   0,43854863 3 2 87 14,4514 6 170 2,73E-13
4 HLEG4   0,402112745 4 2 87 12,11657 8 168 1,33E-13
5 MLEG1   0,372179468 5 2 87 10,61021 10 166 8,52E-14
6 FLEG2   0,345662316 6 2 87 9,57871 12 164 5,83E-14
7 WCUB   0,318459751 7 2 87 8,933568 14 162 2,85E-14
8   WINGLENG 0,332183411 6 2 87 10,04563 12 164 1,3E-14
At each step, the variable that minimizes the overall Wilks' Lambda is entered.
a Maximum number of steps is 34.
b Maximum significance of F to enter is .05.
c Minimum significance of F to remove is .10.
d F level, tolerance, or VIN insufficient for further computation.
Wilks' Lambda
  Number of Variables Lambda df1 df2 df3 Exact F      
Step           Statistic df1 df2 Sig.
1 1 0,718598562 1 2 87 17,034494 2 87 5,72E-07
2 2 0,485050493 2 2 87 18,741188 4 172 8,35E-13
3 3 0,43854863 3 2 87 14,451396 6 170 2,73E-13
4 4 0,402112745 4 2 87 12,116572 8 168 1,33E-13
5 5 0,372179468 5 2 87 10,610209 10 166 8,52E-14
6 6 0,345662316 6 2 87 9,5787098 12 164 5,83E-14
7 7 0,318459751 7 2 87 8,9335681 14 162 2,85E-14
8 6 0,332183411 6 2 87 10,045631 12 164 1,3E-14
Eigenvalues
Function Eigenvalue % of Variance Cumulative % Canonical Correlation
1 1,194928834 76,28274107 76,28274107 0,7378376
2 0,37151833 23,71725893 100 0,5204623
a First 2 canonical discriminant functions were used in the analysis.
Wilks' Lambda
Test of Function(s) Wilks' Lambda Chi-square df Sig.
1 through 2 0,332183411 93,12474771 12 1,223E-14
2 0,729118947 26,69510447 5 6,539E-05
Standardized Canonical Discriminant Function Coefficients
  Function  
  1 2
WCUB 0,577515455 -0,053581519
HINDWING -0,320606939 0,971532131
FLEG1 -1,059788672 -0,18979531
FLEG2 0,548990264 -0,109579618
MLEG1 -0,312188955 0,378993972
HLEG4 0,623071843 0,060961399
Structure Matrix
Function
  1 2
FLEG1 -0,556958903 0,237334618
HINDWING -0,055623155 0,90663887
WINGLENG -0,040946277 0,725462365
HLEG1 -0,03310519 0,618185414
HLEG2 0,027239821 0,614406452
WINGWIEG -0,014961922 0,577449325
HLEG3 0,133027982 0,56035076
MLEG2 0,024676819 0,557261245
MLEG1 -0,138359018 0,524974619
FLEG3 -0,24476504 0,475462636
ANTEN2 0,061467196 0,472444575
HLEG4 0,317086361 0,435456228
MLEG3 0,004271698 0,426116701
WCUB 0,18815847 0,413166533
FLEG2 0,077381935 0,357310448
MLEG4 -0,083465083 0,206400485
ANTEN 0,127415773 0,170152654
Pooled within-groups correlations between discriminating variables and standardized canonical discriminant functions
 Variables ordered by absolute size of correlation within function.
* Largest absolute correlation between each variable and any discriminant function
a This variable not used in the analysis.
Canonical Discriminant Function Coefficients
  Function  
  1 2
WCUB 0,258786368 -0,024010036
HINDWING -0,037088962 0,112390327
FLEG1 -0,196992293 -0,035278933
FLEG2 0,14934115 -0,02980881
MLEG1 -0,073975688 0,089805674
HLEG4 0,140036051 0,013701138
(Constant) 2,80878759 -22,9277581
Unstandardized coefficients
Classification Function Coefficients
  BOLGE    
  malatya ege kayseri
WCUB 0,207219681 -0,205330173 -0,469578025
HINDWING 1,541281403 1,732605509 1,62643057
FLEG1 0,252978015 0,502033488 0,773991595
FLEG2 0,518838116 0,261405457 0,130004793
MLEG1 3,156868778 3,375440036 3,341315273
HLEG4 1,182262428 0,991410693 0,813126501
(Constant) -314,3240831 -345,6645346 -319,102827
Fisher's linear discriminant functions
Classification Results
      Predicted Group Membership     Total
    BOLGE malatya ege kayseri  
Original Count malatya 20 8 2 30
    ege 6 18 6 30
    kayseri 2 6 22 30
  % malatya 66,66666667 26,666667 6,66666667 100
    ege 20 60 20 100
    kayseri 6,666666667 20 73,3333333 100
Cross-validated Count malatya 20 8 2 30
    ege 6 17 7 30
    kayseri 2 6 22 30
  % malatya 66,66666667 26,666667 6,66666667 100
    ege 20 56,666667 23,3333333 100
    kayseri 6,666666667 20 73,3333333 100
a Cross validation is done only for those cases in the analysis. In cross validation, each case is classified by the functions derived from all cases other than that case.
b 66,7% of original grouped cases correctly classified.
c 65,6% of cross-validated grouped cases correctly classified.