Skip to Main Content

Analysis

We constructed the Hassle Factor through confirmatory factor analysis utilizing the statistical software program R (R Core Team, 2019) and applying the lavaan library (Rosseel, 2012). The software package lavaan (Rosseel, 2012) performs structured equation modelling (SEM) and confirmatory factor analysis (CFA) in a manner similar to no-code PLS-SEM platforms such as SmartPLS (Ringle et al, 2015; Rosseel, 2012). R was selected over smartPLS due to its capability for performing several analyses in rapid succession.

The Hassle Factor was computed using 11 ordinal manifest variables (indicators). We conducted the analysis with a diagonally weighted least squares estimator (DWLS). The DWLS estimator is recommended for latent variables constructed using ordinal manifest data as it has been shown to reduce construct bias (Li, 2016). We assumed that the hassle factor impacts the manifest variables outwardly and therefore constructed a reflective model (namely, a high hassle factor is said to influence a manifest variable based on the associated factor loading) (Hair et al, 2017).

Factor loadings can be ranked as they stand for the impact that the hassle factor has on a given manifest variable (Hair et al, 2017). Utilizing the factor loadings we generated for each manifest variable, the countries under study were assigned a hassle factor score computed using latent variable analysis. These rankings are comparable across all 180 countries. Before conducting the analysis, yearly correlation matrices of the factor loadings were produced and used to assess the degree of association among the factor loadings. We found the correlations between the factor variables to be acceptable in all cases (ranging from 0.4-0.8).

References

Fornell, C., & Larcker, D. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50.

Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2016). A primer on partial least squares structural equation modeling (PLS-SEM). Thousand Oaks, CA: SAGE Publications.

Li, C. H. (2016). Confirmatory factor analysis with ordinal data: Comparing robust maximum likelihood and diagonally weighted least squares. Behavior Research Methods 48(3), 936-949. https://doi.org/10.3758/s13428-015-0619-7

Ringle, Christian M., Wende, Sven, Will, Alexander. (2005). SmartPLS 2.0.M3. Hamburg: SmartPLS. Retrieved from http://www.smartpls.com

R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. URL http://www.R-project.org/

Rosseel Y (2012). lavaan: An R Package for Structural Equation Modeling. Journal of Statistical Software, 48(2), 1–36. http://www.jstatsoft.org/v48/i02/.

Schotter, A.,& Beamish P.W. (2013). The Hassle Factor: An Explanation for Managerial Location Shunning. Journal of International Business Studies, 44 (5),521–544.  www.jstor.org/stable/23434160.