Harmonized Latin American Innovation Surveys Database (LAIS): Firm-Level Microdata for the Study of Innovation: 2007-2017

By Competitiveness, Technology and Innovation Division (VPS/PTI/CTI)

To design and promote effective regional innovation policy, we need LAIS—valid, comparable, standardized firm-level innovation survey data across Latin America.
The Harmonized Latin American Innovation Surveys Database (LAIS) contains ~690 variables and ~119,900 firm-level observations from 30 national innovation surveys across 10 Latin American countries (2007–2017).
This dataset expands the region’s publicly available microdata on innovation.

A complementary IDB technical note explains how harmonization was done: selecting variables that measure the same underlying concept across diverse survey methods and instruments.

With LAIS data now available, more researchers can investigate innovation in Latin American firms and address long-standing questions about the factors that drive firms’ innovation decisions.

Show more

Metadata & use

Identifier https://doi.org/10.60966/hbz4-cm10
License Creative Commons Attribution–NonCommercial–NoDerivs 3.0 IGO
Related Knowledge Product
Citation

Crespi, Gustavo, et al. (2022). Harmonized Latin American Innovation Surveys Database (LAIS): Firm-Level Microdata for the Study of Innovation: 2007-2017. IDB Open Data. https://doi.org/10.60966/hbz4-cm10

Published date 2022-03-07
Modified date 2026-07-15
Tags/Keywords Cross-Country Data · Innovation · Innovation Surveys · Microeconomics
Language
  1. English
Temporal coverage 2007-2017
Country
Argentina
Chile
Colombia
Dominican Republic
Ecuador
El Salvador
Panama
Paraguay
Peru
Uruguay
Region Latin America and the Caribbean
Publisher
Inter-American Development Bank
Author
Crespi, Gustavo
Guillard, Charlotte
Salazar, Mónica
Vargas, Fernando
Data collection type Administrative Data
Statistical type Panel Data
Data structure Structured Data
Data notes

What is the temporal and geographic coverage of LAIS?

It covers firm-level data from innovation surveys conducted between 2007 and 2017 in 10 Latin American countries. These surveys are drawn from 30 national innovation surveys.

How many variables and observations are included?

LAIS contains nearly 690 variables and approximately 119,900 firm-level observations.

Which countries are included in LAIS?

LAIS covers 10 Latin American (and Caribbean) countries:
Argentina, Chile, Colombia, Dominican Republic, Ecuador, El Salvador, Panama, Paraguay, Peru, and Uruguay.
The region is labeled as “Latin America and the Caribbean” in the dataset metadata.

What kinds of variables are harmonized across the surveys?

LAIS includes harmonized variables on:
- Innovation activity and expenditures
- Sources of information & collaborations
- Innovation obstacles
- Innovation outputs and effects
- Protection of innovation results
- General firm characteristics

Why was LAIS created — what’s its primary purpose?

The goal is to provide valid, comparable, standardized microdata on innovation to support robust cross-country research and evidence-based regional innovation policy.
It addresses the challenge that innovation surveys across countries use different methods and questionnaires.

What methodological challenges or caveats remain?

Even after harmonization, survey differences remain (e.g., in reference periods, coverage, questionnaire wording).
Comparability is improved but not perfect. Users should consult the technical note for variable-selection criteria and limitations.

What research opportunities does LAIS enable?

LAIS lets researchers explore:
- Firm-level innovation behavior across countries
- Drivers and constraints of innovation in Latin American firms
- Cross-country comparisons to test hypotheses about innovation policy
- The relative importance of internal vs external factors (e.g., collaboration, financing, obstacles) in innovation

Are there examples of research using LAIS?

Yes, for instance:
- A study on how firms innovate in Latin America leverages LAIS to classify innovation strategies via factor & cluster analysis. - Public policy analyses of innovation barriers and cooperation use LAIS data to explore how cooperation can mitigate obstacles.

What does LAIS mean in the context of this dataset?

LAIS stands for Harmonized Latin American Innovation Surveys Database. It is a firm-level microdata resource compiled by the Inter-American Development Bank, covering 30 national innovation surveys conducted between 2007 and 2017 in 10 Latin American countries. It harmonizes nearly 690 variables to allow cross-country comparison of innovation activities.

What kind of information does the LAIS dataset provide?

The dataset includes variables on firm characteristics, innovation expenditures, sources of information, collaboration networks, obstacles to innovation, outputs and effects of innovation, and protection of innovation results. It also contains general firm-level data, including sector classification, employment, exports, and sales.

Which countries are covered by LAIS?

LAIS covers Argentina, Chile, Colombia, Ecuador, El Salvador, Dominican Republic, Panama, Paraguay, Peru, and Uruguay.

What is the time period of the surveys included in LAIS?

The harmonized dataset spans surveys conducted between 2007 and 2017, providing a decade of comparable firm-level innovation data across Latin America.

How many observations and variables are in the LAIS dataset?

LAIS contains approximately 119,900 firm-level observations and 687–690 harmonized variables.

What is the purpose of harmonizing these innovation surveys?

Harmonization ensures that variables measuring the same underlying concepts across different national surveys are standardized. This allows researchers and policymakers to conduct regional analyses, compare innovation dynamics across countries, and design evidence-based innovation policies.

Who created and maintains the LAIS dataset?

The dataset was developed by the Inter-American Development Bank’s Competitiveness, Technology, and Innovation Division. Key contributors include Gustavo Crespi, Charlotte Guillard, Mónica Salazar, and Fernando Vargas.

Dataset files

Load more

Additional materials

Load more