The Effects of Public Institution Characteristics on Exporting SMEs

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

20

Table 1. Frequencies

Frequencies Age Employees Internationalization

(21)

(22)

22 variance extracted (AVE). Accordingly, the threshold of 0,7 for CR and 0,5 (Hair, 1979) for AVE is passed, except for formalization (CR=0,682 and AVE=0,420). This means that the relation within formalization is low, which may be due to the lower content validity. Other reasons would be the small sample size (n = 79) or because the respondents were answering from their public partner perspective. This will be further elaborated on in the discussion. The convergent validity is completely past for political directives, centralization and knowledge acquisition. For formalization, it is partially reached since AVE is smaller than 0,5. With respect to the discriminant validity, the entire model is confirmed (MSV < AVE and ASV < AVE). Table 2 shows the construct, convergent and discriminant validity. Finally, a reliability analysis is done and the Cronbach Alpha’s were evaluated. The α-value for political directives is 0,967 and can be considered very high. Accordingly, the α-value for centralization is 0,850 and for knowledge acquisition it is 0,791. Hereby, all values are strongly correlated which makes the items sufficient (>0,6). However, the α-value of formalization is 0,686, which is slightly lower but still considered acceptable (Šebjan and Tominc, 2014). Accordingly, the scales equip high internal consistency and are therefore reliable for further analysis. The items of the final scales with their; Cronbach alpha’s, factor loadings, eigenvalues and variance explained are presented in Table 3. Table 2. Convergent and discriminant validity of measurement models

Construct CR AVE MSV ASV Results

(23)

23

Table 3. Factor analysis: EFA, Varimax

Construct Questions Factor loadings

(24)

(25)

(26)

26

Table 5. Pearson correlations

Variables KA For Cen PD Age

SMEs (Log) Empl. Int. Export %

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41)

41 9.2. APPENDIX B: METHODOLOGHY

Table B-1. Total Variance Explained Component Initial Eigenvalues Rotation Sums of Squared Loadings

Total % of Variance Cumulative % Total % of Variance Cumulative %

(42)

42

Figure B-1. AMOS

Table B-3. Model Fit Summary CFA

CMIN

Model

NPAR

CMIN DF

P

CMIN/DF

Default model

35 83.066 70 .136

1.187 Saturated model

105 .000

0 Independence model

14 817.963 91 .000

8.989 Baseline Comparisons

Model

RFI

_{IFI TLI CFI}

Delta1 rho1 Delta2 rho2

Default model

.898 .868

.983 .977

.982

Saturated model

1.000

1.000 Independence model

.000 .000

.000

RMSEA

Model

RMSEA

LO 90 HI 90

PCLOSE

Default model

.049

.000

.086

.495

(43)

43 Threshold: *CMIN/DF<2,0 **CFI<0,9 ***RMSEA<0,8 Table B-4. Validity Table

CR AVE MSV MaxR(H) KA PD Cen For

(44)

44 9.3. APPENDIX C: RESULTS

Figure C-1. Scatterplot Knowledge Acquisition

Table C-1. Test of Normality

Kolmogorov-Smirnova _Shapiro-Wilk

Statistic df Sig. Statistic df Sig.

PD .198 79 .000 .923 79 .000 Cen .148 79 .000 .957 79 .010 KA .123 79 .005 .968 79 .047 For .123 79 .005 .966 79 .034 a. Lilliefors Significance Correction

Table C-2. Residuals Statistics

Minimum Maximum Mean Std. Deviation N

(45)

45

Table C-3. Descriptive Statistics

Knowledge Acquisition Formalization Centralization Political Directives

Statistic Std. Error Statistic Std. Error Statistic Std. Error Statistic Std. Error

(46)

46

Table C-4. Regression Model 1

Model Summaryb

Model R R Square Adjusted R Square

Std. Error of the Estimate 1 .250a _.062 _.025 _.99262 a. Predictors: (Constant), For, PD, Cen b. Dependent Variable: KA

ANOVAa

Model Sum of Squares df Mean Square F Sig.

1 Regression 4.919 3 1.640 1.664 .182b Residual 73.897 75 .985 Total 78.816 78 a. Dependent Variable: KA b. Predictors: (Constant), For, PD, Cen

Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics

B Std. Error Beta Tolerance VIF

1 (Constant) 4.125 .664 6.216 .000 PD -.133 .115 -.141 -1.154 .252 .840 1.190 Cen -.042 .165 -.032 -.255 .799 .783 1.277 For -.220 .174 -.153 -1.265 .210 .853 1.172 a. Dependent Variable: KA Table C-5. Regression Model 2 Model Summaryc Model R R Square Adjusted R Square Std. Error of the Estimate Change Statistics R Square

Change F Change df1 df2 Sig. F Change

(47)

47

ANOVAa

Model Sum of Squares df Mean Square F Sig.

1 Regression 4.919 3 1.640 1.664 .182b Residual 73.897 75 .985 Total 78.816 78 2 Regression 6.653 7 .950 .935 .485c Residual 72.163 71 1.016 Total 78.816 78 a. Dependent Variable: KA b. Predictors: (Constant), For, PD, Cen c. Predictors: (Constant), For, PD, Cen, Int, Log_Empl, Age, ExportPerc Table C-6. Coefficients Regression Model 1 & 2 Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics

B Std. Error Beta Tolerance VIF

(48)

(49)

(50)

(51)

(52)