Multiple Regression Analysis was used to determine whether the eight independent variables, which are SMEs
characteristic, management and know-how, products and services, Customer and Market, the way of doing
business and cooperation, resources and finance, Strategy, and external environment, have any significant effect
toward Business Success of SMEs in Thailand. The results are shown in Table 3.
The findings of the study revealed that SMEs Characteristic, Customer and Markets, the way of doing Business
& Cooperation, Resources and Finance, and External Environment have significant positive effect on the
Business Success of SMEs in Thailand. Management know-how, Product and Services, and Strategy were found
to have no significant effect on the Business Success of SMEs in Thailand.
The overall results of the regression analysis shows that this model is well constructed and it is well represented
as reflected in the variables selected. Table 2, the summary table on regression analysis indicated that the
R-square is 53.3 percent. This means that the eight variables which include SMEs characteristic, management
and know-how, products and services, customer and markets, the way of doing business and cooperation,
resources and finance, strategy, and external environment can explain 53.3 percent variations in the business
success of SMEs in Thailand.
The Durbin-Watson statistic shows that the serial correlation of residuals is 1.842, the value falls within the
acceptance range. (1.5 and 2.5). This means that there is no auto correlation problem in the data. The Condition
Index, Variance Inflation Factors (VIF) and tolerance all fall within the acceptance range (Condition index =
27.544, VIF = 1 - 10, tolerance = 0.1 – 1.0). This means that there is no multi-co linearity problem in the
regression model used for this study. The histogram indicates that data used in this study is normally distributed
and F-value is found to be significant at 1% significance level (sig. F = .000). This concludes that the regression
model used in this study is adequate or in other words, the model was fit.