*4.3. Research Model*

We use the OLS model for empirical analysis. First, to examine the impact of TMT experience heterogeneity on innovation quality, the model is designed as follows:

$$\text{Eiq} = \beta\_0 + \beta\_1 \text{EH} + \beta(\text{Control} + \text{Year} + \text{Industry}) + \varepsilon \tag{1}$$

In Model (1), Eiq is enterprise innovation quality. EH is experience heterogeneity including FEH and IEH. FEH and IEH are respectively substituted into the formula for calculation.

Second, to examine the mediating effect of enterprise partner diversity (Epd), we also use the OLS regression model. The model is designed as follows:

$$\text{Eiq} = \beta\_0 + \beta\_1 \text{EH} + \beta\_2 \text{Epd} + \beta(\text{Control} + \text{Year} + \text{Industry}) + \varepsilon \tag{2}$$

Epd is added on the basis of Model (1). In Model (2), if the coefficient of β<sup>2</sup> is positive, it indicates that partner diversity plays an intermediary role in promoting enterprise innovation. Based on Hypotheses 3 and 4, we expect β<sup>2</sup> to be significantly positive. The definition of the remaining variables in Model (2) is the same as in Model (1).

Third, to test the moderating effect of TMT technological participation (TMTTP), we add the variable TMTTP and its interaction term with the Epd to Model (1). The model is designed as follows:

$$\text{Epd} = \beta\_0 + \beta\_1 \text{EH} + \beta\_2 \text{TMTP} + \beta\_3 \text{Epd} \ast \text{TMTP} + \beta(\text{Control} + \text{Year} + \text{Industry}) + \varepsilon \tag{3}$$

FEH and IEH are respectively substituted into the formula for calculation, and the other variables are defined the same as the previous models. In Model (3), if the coefficient of β<sup>3</sup> is positive, it indicates that TMT technological participation positively moderates the relationship between TMT experience heterogeneity and partner diversity. Based on hypotheses 5 and 6, we expect β<sup>3</sup> to be significantly positive.
