异质性、教育发展与国家创新能力——基于面板分位数模型的研究

Heterogeneity,Education Development and National Innovation Ability——Based on Panel Quantile Model

建立固定效应模型和面板分位数模型,发现后者能够更深入地揭示不同创新水平下教育发展对创新能力的影响。使用PISA测试成绩建模的结果显示,阅读成绩对创新指数有重要的正向影响,创新能力对其的弹性约为2.49,即阅读成绩提高1%将使创新指数提高2.49%;数学成绩和科学成绩对创新指数的影响为负但基本不显著。平均受教育年限对创新指数有显著的正效应,创新能力对其的弹性约为0.38,即平均受教育年限提高1%将有助于创新指数提高0.38%,而且这一影响在创新能力的不同层次下保持稳定。根据所选国家是否为OECD成员国进行分组,对OECD国家数据重新建模,得到了与全样本一致的结果,证明结论具有稳健性。研究表明,阅读能力、平均受教育年限等教育发展指标对提升创新能力有显著的促进作用。

By respectively building fixed effect model and panel quantile model to analyze the relationship between innovation ability and education development, it is found that the panel quantile model can deeply reflect the effect of education development on innovation ability under different levels of innovation. The scores of PISA test are taken as independent variables in the panel quantile model, the results show that there is an important positive impact of the score of reading part to innovation index, the elasticity that innovation ability to this variable is 2.49, namely, if there is 1% improvement on the score of reading part, the innovation index will be improved about 2.49%. The score of mathematics and science part has negative influences on innovation index, but lots of coefficients of these variables are insignificant. The results of panel quantile model show that there is a significant positive effect of average years of total schooling on innovation index. The elasticity that innovation ability to this variable is 0.38, namely, if there is 1% improvement on the average years of total schooling, the innovation index will be enhanced about 0.38% and this result is stable under different innovation ability levels. This paper splits the initial sample to two parts which represent OECD countries and non-OECD countries, respectively. The conclusion of the panel quantile model about the OECD sample in line with above results and then attests the model robustness. This research shows that education development indicators, such as reading ability, average years of total schooling, and so on, are beneficial to the improvement of innovation ability.