In order to analyze the organizational performance of government-funded research institutes, four independent variables (organization size, organization age, evaluation system, organization location) and goal ambiguity were empirically analyzed. Although there are existing studies on institutional evaluation, which is the organizational performance of government-funded research institutes, there are differences in research in that it is an empirical analysis analyzed from the perspective of goal ambiguity and that the evaluation system, which is the specificity of institutional evaluation, is included in the analysis. As a result of the analysis, organizational location was the only independent variable that had a direct significant effect on organizational performance, and goal ambiguity showed a significant mediating effect for all independent variables. This indicates that goal ambiguity plays an important role in the organizational performance of government-funded research institutes, and to reduce this, increase the size of the organization, pay attention to establishment of the organization for organizational age, and that it is desirable to have an independent research group in the government department for the organization location. In addition, it was supported that the improvement effect of the evaluation system is proceeding in a desirable direction in terms of goal ambiguity.

The effects of R&D investment on innovation such as patents and intangible assets, and the effect on the corporate performance such as revenue and profit growth, were analyzed using path analysis. In particular, this study compared and analyzed the performances of non-cooperative R&D and cooperative R&D. The results of this study are summarized as follows. First, R&D investment has a significant impact on innovation performance. This supports the existing research results. Second, patents have a significant impact on intangible asset growth. Third, in the case of corporate groups carrying out cooperative R&D, intangible asset growth forms a significant causal relationship with revenue growth. Fourth, in case of cooperative R&D, intangible asset growth has a significant mediating effect between patent and revenue growth. Like the existing research, the results of this research support the innovation performance of R&D investment. It also supports the existing argument that the results of cooperative R&D are more favorable to increase corporate value. However, unlike the existing research, we found a path leading to increased revenue through patents and intangible assets, and confirmed that such a path is likely to be achieved through cooperative R & D rather than internal R&D.

The growth and employment effects of R&D investment were analyzed according to business size, export value and manufacturing sectors so as to suggest improvement directions for effective industry policies. The effect of R&D investment was considered simultaneously from the two perspectives of growth and employment effect, and the causality analysis was carried out by using a path analysis. The result of the path analysis confirmed significant differences in the growth effect of R&D investment depending on business size. However, the effect of increasing employment was difficult to obtain statistically significant results for any various combinations of business size and export value. This is a mixture of directions for the effects of R&D investment on employment, which could be due to the failure to consider appropriate time lags between investment and effect. Efficiency analysis by industry sectors confirmed significant differences in efficiency depending on business size, but differences depending on export value were difficult to identify. In order to derive improvement policy by industry sector according to business size and export value, the direction of selective support policy and universal support policy was derived for six industry groups by combining the return to scale in the efficiency analysis and R&D concentration. Hirschman-Herfindahl index is used for calculating R&D concentration.