Media and Economic Growth: Country Groups Difference via Panel Data
Viktoriia Ahapova
Abstract
Open-access reader
Viktoriia Ahapova
Abstract
Open-access reader
The present article investigates the link between economic growth, namely GDP per capita, and the media activity represented with the indicator of the press freedom alongside other factors such as infrastructure, institutional conditions, and foreign direct investments. A panel of 179 countries was used for the period from 2000 to 2015. In particular, we run two panel data analysis models, fixed effects and random effects models, and examined their output with Hausman’s specification test, which pointed the fixed effects model as more efficient for the presented data set. However due to the presence of serial correlation, heteroskedastic, and cross-panel dependence, a Prais-Winsten regression with panel corrected standard errors (PCSE) was implemented. The comparative analysis of models of four country groups, divided by GNI per capita, was conducted. Both statistically significant correlation coefficients and models’ output provided evidence of an association between economic growth and the press activity.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The present article investigates the link between economic growth, namely GDP per capita, and the media activity represented with the indicator of the press freedom alongside other factors such as infrastructure, institutional conditions, and foreign direct investments. A panel of 179 countries was used for the period from 2000 to 2015. In particular, we run two panel data analysis models, fixed effects and random effects models, and examined their output with Hausman’s specification test, which pointed the fixed effects model as more efficient for the presented data set. However due to the presence of serial correlation, heteroskedastic, and cross-panel dependence, a Prais-Winsten regression with panel corrected standard errors (PCSE) was implemented. The comparative analysis of models of four country groups, divided by GNI per capita, was conducted. Both statistically significant correlation coefficients and models’ output provided evidence of an association between economic growth and the press activity.
Key concepts: Panel data, Hausman test, Econometrics, Fixed effects model, Random effects model, Per capita, Heteroscedasticity, Economics