Reprodution Folder"Kinship Intensity and Innovation Ambidexterity in Family Businesses: Venturing into the Unknown"

This dataset contains the replication materials for the article "Kinship Intensity and Innovation Ambidexterity in Family Businesses: Venturing into the Unknown" submitted in Economic Modelling.
The sample consists of 7,989 firm-year observations of Chinese A-share listed family businesses from 2012 to 2020. The data are primarily sourced from the China Stock Market & Accounting Research (CSMAR) database, with supplementary manual collection for kinship relationship information.
Key variables include:

Exploratory innovation (Expr): Measured by the number of exploratory patent applications (invention patents and design patents) plus one, followed by a natural logarithm transformation. Source: CSMAR patent database.
Exploitative innovation (Expi): Measured by the number of exploitative patent applications (utility model patents) plus one, followed by a natural logarithm transformation. Source: CSMAR patent database.
Kinship intensity (Kinship): The primary independent variable, calculated as the sum of kinship coefficients between the de facto controller and all family members on the board of directors. Kinship coefficients follow standard genealogical weighting (e.g., 0.5 for parent-child or spouses, 0.25 for siblings or grandchildren, etc.).
Control variables: Firm size, age, leverage, ROA, cash flow, board size, independent director ratio, ownership concentration, industry and year dummies, and other standard firm-level controls (full list available in the manuscript).
Instrumental variable: Per capita rice cultivation area in 1978 in the birthplace of the ultimate controller (used for IV estimation).

Important Privacy Notice:
The variables related to kinship relationships (e.g., Kinship, Kinship_core, Kinship_noncore, Kinship_blood, Kinship_inlaw, and family member identifiers) were manually collected from multiple sources including annual reports, prospectuses, listing announcements, executive supervision reports, news articles, and other public and semi-public channels. These data involve detailed familial relationships of entrepreneurs and controlling families, which are considered highly sensitive personal information.
Due to privacy and ethical considerations, researchers interested in accessing or replicating results using the kinship variables are kindly requested to contact the corresponding author or the first author directly for collaboration or controlled access under appropriate data use agreements.
The remaining variables (financial, patent, governance, and instrumental variable data) are derived from public sources (primarily CSMAR) and can be independently reconstructed by licensed users of the database using the variable definitions and sample selection criteria provided in the manuscript.