A Framework to Assess the Citation Performance of Complex Innovation Systems

Area of Science:

• Bibliometrics and Scientometrics
• Innovation Studies
• Science Policy Evaluation

Background:

• Scientometric indicators are vital for research evaluation, rankings, and policy-making.
• Selecting appropriate indicators is critical for accurate assessment.
• Complex innovation systems require specialized evaluation methods.

Purpose of the Study:

• To introduce a framework for selecting appropriate scientometric indicators for complex innovation systems.
• To provide a methodology for deciding between standard average-based and scale-invariant indicators.
• To offer procedures for constructing scale-invariant indicators to assess relative impact.

Main Methods:

• Development of a two-fold framework for indicator selection.
• Methodology to determine the necessity of scale-invariant indicators.
• Empirical validation using the Chilean science system's 2017 data.

Main Results:

• The Chilean science system exhibits characteristics of a complex innovation system.
• Citation distribution fits a power law with an exponential cutoff.
• A power-law correlation exists between system size and impact.
• The framework efficiently compares fields of varying sizes.

Conclusions:

• The proposed framework effectively guides the selection of scientometric indicators for complex innovation systems.
• Scale-invariant indicators are essential for accurately assessing the relative impact of such systems.
• The framework facilitates robust comparisons across diverse scientific fields.