*3.2. Definition of Variables and Statistical Techniques*

CSMAR uses absolute value and relative value to quantify the investment in research and development. The absolute value refers to the total amount of R&D investment within a certain time range, while relative value refers to the ratio between the total amount of R&D investment and the total sales [38–40], total assets [41–43], or the employees' number [44,45] within a certain time range. This study mainly focuses on the relative degree of attention paid to innovation for research and development by listed companies. Taking into account the availability of data, and the various sizes of companies, this project operationalized R&D investment as the proportion of total amount of R&D investment normalized by total sales.

This paper examines the postgraduate education of directors from the perspectives of education level, discipline background, and graduation institution. This paper uses the proportion of directors with master's degrees and proportion of directors with doctoral degrees as the two indicators of education level. Discipline background is quantified using: the proportion of directors with degrees from science and engineering field and the proportion of directors with degrees from economics and management. Graduation institution of directors is measured upon: the percentage of directors graduated from C9 universities and the percentage of directors graduated from overseas institutions. The C9 is China's first collegiate league among elite universities. It was launched in October 2009. The alliance members include Peking University, Tsinghua University, Fudan University, Shanghai Jiaotong University, Nanjing University, Zhejiang University, University of Science and Technology of China, Harbin Institute of Technology, and Xi 'an Jiaotong University. The postgraduate education characteristics of directors is not the only potential factor of R&D investment. Directors' working experience, corporate asset-liability ratio, industry type, and total assets are adopted as control variables in this study.

All the variables in this paper and its definition are listed in the Table 2.


#### **Table 2.** Variable definition.

Note: CC: commercial companies; FR: financial and real estate companies; PU: public utility companies; AFAF: companies in agriculture, forestry, animal husbandry, and fishery; IC: industrial companies; CES: cultural, educational, and sports companies; and IT: IT companies.

This paper uses basic descriptive statistics to explore the data first. OLS regression is used to test the influence of directors' postgraduate education characteristics on the firm's R&D investment overall and by industry. Taking into account potential endogeneity issues, more control variables were added and IV2SLS regression was employed to ensure the robustness of the result.

#### *3.3. Descriptive Statistics*

Table 3 shows the descriptive statistics for all variables of interest in this paper. In 2017, the average proportion of R&D investment in listed companies of China reached 5.10% and the maximum proportion of R&D investment reached 76.35%. In terms of directors' postgraduate education level, nearly half (48%) of directors in listed companies have master's degrees and 24% have doctoral degrees as their highest degrees. In terms of discipline background, approximately 46% of directors got degrees from economics and management and 15% from science and engineering. In terms of graduation institutions, about 18% of directors graduated from C9 institutions in China and 15% from overseas institutions. In addition, roughly 24% of directors were engaged in research and development positions once.


**Table 3.** Results of descriptive statistics for each variable.

R&D investment amount varies by industry. As shown in Figure 1, the selected companies on average invested about 5.10% of the total sales on R&D activities. The IT industry has the highest (10.57%) R&D investment ratio among all industries, followed by industrial companies (4.74%) and cultural, educational, and sports companies (4.32%). Commercial companies spent the least (0.92%) among all industries on R&D activities.

**Figure 1.** Distribution of R&D investment for listed companies by industry.

Table 4 presents the distribution of directors' postgraduate education characteristics in listed companies by industry. In term of education level, the proportion of directors with master's degrees in listed companies is higher than that with doctoral degrees across industries. The largest gap of this exists in agriculture, forestry, animal husbandry, and fishery companies: where the proportion of directors with doctoral degrees is the lowest and the proportion of directors with master's degrees is the highest among all industries. The proportion of directors with doctoral degrees in public utility companies is the highest. In term of discipline background, the proportion of directors majoring in economics and management in each industry is far higher than that in science and engineering, which is most prominent in agriculture, forestry, animal husbandry, and fishery companies. In term of graduation institution, the percentage of directors graduating from C9 universities and overseas institutions in agriculture, forestry, animal husbandry, and fishery companies is the highest. The lowest figure for directors graduating from C9 universities and graduating from overseas institutions appears in industrial and commercial companies, respectively.


**Table 4.** Distribution of directors' postgraduate education characteristics in listed companies by industry.

Note: CC: commercial companies; FR: financial and real estate companies; PU: public utility companies; AFAF: companies in agriculture, forestry, animal husbandry, and fishery; IC: industrial companies; CES: cultural, educational, and sports companies; and IT: IT companies.

#### **4. Results**
