Validation of Measures

3928 Words8 Pages

4 Analyses and results

4.1 Validation of measures

This study refines the measures and assesses the reliability and validity. First, the values of Cronbach’s alphas for three components of market orientation, incremental and radical service innovations, market and financial performances, market turbulence, technological turbulence, and competitive intensity range from 0.70 to 0.92, all of which are greater than the recommended level of 0.70 (Nunnally, 1978). Based on the first cut, the internal consistency and reliability of the four scales were established. Next, to understand the factor structure and the measurement quality, a principal component analysis was conducted with varimax rotation. An evaluation of the eigenvalues was used to identify the number of factors to retain. Throughout this process, factors are loaded as expected. Thus, these results indicate the unidimensionality of the various constructs.

Then, this study estimated a 12-factor confirmatory measurement model, including 10 focal constructs and two controls. For the multi-item scales, this study set each questionnaire item to load only on its respective latent construct, and allowed the latent constructs to correlate. After dropping one item that possessed low factor loading, the confirmatory model fit the data satisfactorily (X / d.f. = 1.32, RMSEA= 0.05, CFI = 0.92, NNFI = 0.95, PNFI = 0.76). All factor loadings are statistically significant (from 0.62 to 0.91) as they are higher than the recommended level of 0.60 (Hair et al., 2006). The t-value of each factor loading is also significant (p < .01). The composite reliabilities of constructs exceed the usual 0.60 benchmark, ranging from 0.80 to 0.91 (Bagozzi and Yi, 1988), and all of the extracted average v...

... middle of paper ...

... could potentially overcome such biases. Third, because the sample for this study is drawn from a wide range of service sectors, the applicability of these findings to other industries should be considered with caution. Fourth, this study focuses on the mediating role of the market orientation and performance relationship. Further research could examine the boundary conditions of market orientation. For example, the role of customer orientation may be particularly salient in interaction needed service sectors while competitor orientation may be significant in self-service technologies needed areas. Finally, this study was conducted in Taiwan where there may be a higher inclination toward the service industry than many other countries. Replication of this research in other countries should help to verify the applicability of the results for other parts of the world.

Open Document