Multivariate Exploratory Data Analysis for Large Databases: An Application to Modelling Firms' Innovation using CIS Data. Issue 4 (October 2019)

• Record Type: Journal Article
• Authors: Bou, Juan C.; Satorra, Albert
• Is Part Of: Business research quarterly. Volume 22:Issue 4(2019)
• Publication Date: 2019-10
• Subjects: Community Innovation Survey (CIS) -- MEDA -- Innovation -- Missing data -- MAR and MCAR -- Dimension reduction -- Multivariate analysis -- OLS -- ordered logistic and Tobit regression; 658
• DOI: 10.1016/j.brq.2018.10.001
• Languages: English
• ISSNs: 2340-9436

Abstract

This paper argues that, when using a large database, organizational researchers would benefit from the use of specific multivariate exploratory data analysis (MEDA) before performing statistical modelling. Issues such as the representativeness of the database across domains (countries or sectors), assessment of confounding among categorical covariates, missing data, dimension reduction to produce performance indicators and/or remedy multi-collinearity problems are addressed by specific MEDA. The proposed MEDA is applied to data from the Community Innovation Survey (CIS), a large database commonly used to analyse firms' innovation activities, prior to fitting ordered logit and Tobit regression models. A set of recommended practices involving MEDA are proposed throughout the paper.