Home Explore Example 03-1- Tsai, Yang, Firm Innovativeness and Performance

# Example 03-1- Tsai, Yang, Firm Innovativeness and Performance

## Description: Example 03-1- Tsai, Yang, Firm Innovativeness and Performance

1286 K.-H. Tsai, S.-Y. Yang / Industrial Marketing Management 42 (2013) 1279–1294 Table 2 Descriptive statistics, correlation matrix, and reliabilities.a Variables Mean S. d. 1 2 3 4 5 6 1 Firm size 3.10 1.35 1.33 2 Firm age 3.51 0.82 0.01 1.09 0.18⁎ 3 Firm innovativeness 5.38 1.33 0.21⁎⁎ − 0.13 0.77 0.34⁎⁎⁎ 0.25⁎⁎ 0.19⁎ 1.17 0.18⁎ − 0.01 0.33⁎⁎⁎ 0.87 0.03 0.002 4 Business performance 4.34 0.16⁎ 0.22⁎⁎ 0.04 0.79 0.56⁎⁎⁎ 0.09 0.16⁎ 0.01 0.44⁎⁎⁎ 0.70 5 Market turbulence 4.76 0.05 0.92 0.76 0.79 0.87 0.92 0.76 0.79 6 Competitive intensity 4.80 0.88 Cronbach's α CR Notes: The square roots of AVE are boldfaced in the diagonal. The correlations (φ) between any two constructs (PHI in the CFA model) are in the upper-right of the diagonal. The Pearson correlations between any two constructs (unweighted mean of the items for each construct) are in the lower-left of the diagonal. a n = 154. ⁎ p b 0.05 (two-tailed test). ⁎⁎ p b 0.01 (two-tailed test). ⁎⁎⁎ p b 0.001 (two-tailed test). 4.3. The results of hypothesis testing this study ﬁrst examined the results of Hypothesis 4, the three-way interaction (the highest order term in model 4). As Table 3 shows, This study employed hierarchical moderated regression analysis the increase in R2 from model 3 to model 4 is .03, and it is statistically to test the hypotheses. This study established four regression models, signiﬁcant (ΔR2 = .03, F change = 5.49, p b .05, two-tailed test), evaluated the incremental variance explained (ΔR2), and conducted which indicates that in model 4, the addition of the three-way inter- overall and incremental F tests of statistical signiﬁcance. This study action among ﬁrm innovativeness, market turbulence, and competi- entered the control variables into the regression equation in step 1 tive intensity signiﬁcantly increased 3% of the explanation of (model 1), all three predictors in step 2 (model 2), three two-way variance in performance (the explanatory power of the model). Addi- interactions in step 3 (model 3), and the three-way interaction in tionally, model 4 in Table 3 shows that the unstandardized coefﬁcient step 4 (model 4). Model 4 includes all of the predictors, the two- for the three-way interaction effect was positive and signiﬁcant (b = way interactions, and the three-way interaction. Table 3 shows the .15, t = 2.34, p b .05, one- and two-tailed tests). These ﬁndings pro- results of the regression analyses (models 1 to 4). As models 2 to 4 vide support for Hypothesis 4. of Table 3 show, ﬁrm innovativeness had a positive and signiﬁcant main effect on business performance (all b > 0 at p b .001). This ﬁnd- To further assess the nature of the three-way interaction, this study ing supports Hypothesis 1, corresponds to the ﬁndings of some em- calculated the slopes of the simple regression equations of business pirical studies (e.g., Hult et al., 2004; Rhee et al., 2010; Theoharakis performance on ﬁrm innovativeness for each of the four conﬁgurations & Hooley, 2008), and validates Barney's (1986) conceptual argument of high and low levels of market turbulence (MT) and competitive that innovativeness can have positive economic value for ﬁrms. Next, intensity (CI) (i.e., one standard deviation above and below their this study tested the moderation hypotheses. Following previous respective means) and tested whether each simple slope was signiﬁ- studies (e.g., Hekman et al., 2009; Shalley, Gilson, & Blum, 2009), cantly different from zero (two-tailed tests) (Aiken & West, 1991). The results of the slope tests indicated that the performance effect of Table 3 Results of hierarchical moderated regression analysis (dependent variable: business performance).a Variables Model 1 Model 2 Model 3 Model 4 VIFs Constant 4.68 (2.77) 4.72 (2.31) 4.65 (2.20) 4.68 (2.30) 1.09 Controls 0.18⁎⁎ (− 0.16) (0.49) (0.31) (0.39) 1.05 − 0.01 (− 1.14) 0.15⁎ (− 1.32) 0.14⁎ (− 1.40) 0.14⁎ (− 1.38) 1.03 Firm size − 0.37 0.03 0.02 0.02 Firm age − 0.41 (4.11) − 0.43 (4.24) − 0.42 (3.43) 1.24 IND 2.80⁎ 0.44⁎⁎⁎ (− 0.58) (− 0.84) (− 0.96) 1.47 Main effects 0.05 − 0.04 (− 0.60) 0.45⁎⁎⁎ (− 0.11) 0.38⁎⁎⁎ (− 0.28) 1.43 Firm innovativeness (FI) 0.03⁎ − 0.05 − 0.06 − 0.07 Market turbulence (MT) 0.05⁎ − 0.01 (1.99) − 0.02 (2.53) 1.40 Competitive intensity (CI) 2.80⁎ 4.36⁎⁎⁎ (− 0.36) (0.07) 1.37 Two-way interactions 0.15 0.18⁎ 0.23⁎ (− 0.18) 1.78 FI × MT 0.12⁎⁎⁎ − 0.04 (1.64) 0.01 FI × CI 0.10⁎ − 0.01 (2.34) 2.08 MT × CI 5.67⁎ 0.08 Three-way interaction 0.15⁎ FI × MT × CI 3.68⁎⁎⁎ 3.97⁎⁎⁎ F 0.19 0.22 R2 0.14 0.16⁎ Adjusted R2 0.04+ 0.03⁎ ΔR2 2.13+ 5.49⁎ F change for ΔR2 IND (industry dummy): 1 = ICT industry; 0 = non-ICT-industry. Unstandardized regression coefﬁcients are reported; t-values are in parentheses. ΔR2 means the increase in R2 from the model to the previous model. a n = 154. + p b .10 (two-tailed incremental F-test for ΔR2). ⁎ p b .05 (one- and two-tailed tests for hypotheses; two-tailed test for control variables). ⁎⁎ p b .01 (two-tailed test for control variables). ⁎⁎⁎ p b .001 (one-tailed test for main effects).