Menu
Home
Log in / Register
 
Home arrow Business & Finance arrow Proceedings of the 1st AAGBS International Conference on Business Management 2014
< Prev   CONTENTS   Next >

E. Assessment of Goodness of Measures

To assess the reliability of the factors using PLS loading factors, the criterion of

0.50 recommended by Krieg [21, 22] was adopted for retention on the factors. When factor loading is closely examined, two items from construct firm characteristic, namely, firm regional and firm status, one item from construct innovation activities and one item from construct TI, were reported with substandard factor loadings (<0.5). All these four factors were dropped from further analysis (see Fig. 1.1). The final model for further analysis is presented in Fig. 1.2. The objective of this study is to identify the factors influencing the technology innovation adoption by firm size. Statistical measurements are applied to identify the significant relationship between the variables. There are two main criteria used for testing goodness of measures, namely, validity and reliability.

F. Construct Validity

Construct validity testifies to how well the results obtained from the use of the measure fit the theories around which test is designed [23]. To construct validity, factor loading is applied to assess if there are problems with any particular items. As suggested by [24], a cutoff value for loading at 0.5 as significant and dropping out insignificant factors having factor loadings of less than 0.5. Any item with a loading of 0.5 and higher on two or more factors would be a deemed significant cross loading. Table 1.1 shows that all the items measuring a particular construct loaded highly on that construct and loaded lower on the other constructs thus confirming construct validity [25].

Fig. 1.2 Result of path analysis

Table 1.1 Loadings and cross loading

Measurement item

ET

FC

IA

TI

ET_FORMAL

0.787

-0.440

-0.085

0.312

ET_HRDF

0.811

-0.498

-0.073

0.322

ET_INDOOR

0.771

-0.425

-0.066

0.335

ET_SKILL

0.585

-0.339

-0.069

0.295

ET_TR1

0.851

-0.439

-0.122

0.382

FC_FIRMSIZE

-0.647

0.799

0.071

-0.383

FC_SW01

-0.263

0.813

0.013

-0.199

FC_SW02

-0.266

0.815

0.011

-0.200

INNO_NP

-0.073

0.023

0.870

-0.217

INN_EXNP

-0.119

0.066

0.874

-0.221

TI_01

0.386

-0.322

-0.125

0.789

TI_02

0.302

-0.216

-0.205

0.766

TI_04

0.300

-0.254

-0.281

0.734

TI_05

0.394

-0.337

-0.187

0.848

TI_06

0.271

-0.240

-0.175

0.716

Bold values are loadings for items which are above the recommended value of 0.5

G. Convergent Validity

Results indicated that the construct ranged from 0.850 to 0.880 which exceeded the recommended value of 0.7 [24], and the variance extracted for four scales ranged from 0.595 to 0.761 which is higher than 0.5 (Table 1.2).

Table 1.3 summarizes the results of measurement model. The results show that all the four constructs, namely, employee training, firm characteristic, innovation activities, and TI, are all valid measures of their respective constructs based on their parameter estimates and statistical significance [24].

 
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Philosophy
Political science
Psychology
Religion
Sociology
Travel