Cultural values change over time due to economic growth and generational shifts. According to the Modernization Theory proposed by Ronald Inglehart and Christian Welzel, as societies transition from agrarian to industrial and post-industrial economies, their moral landscape undergoes a fundamental shift.
This shift typically occurs along two major axes:
While the theory is well-established, visualizing this “cultural drift” over time is statistically complex. A simple comparison of means ignores the multidimensional structure of culture. Moreover, constructing separate maps (e.g., via PCA or MDS) for different time periods often results in unrelated coordinate systems, making direct comparison impossible due to arbitrary rotations.
To address this, this study employs a rigorous geometric approach combining Multidimensional Scaling (MDS) with Procrustes Analysis.
The primary objective of this research is to map the trajectory of global cultural change between two distinct epochs:
By aligning the cultural “constellations” of these two periods, we aim to visualize the “arrows of change,” identifying which nations have undergone the most profound moral transformations regarding religion, trust, and post-materialism.
The empirical analysis was conducted using the Quality of Government (QoG) Standard Dataset (January 2025 version). The primary source of the cultural indicators utilized in this study is the World Values Survey (WVS), which provides comparative data on changing values and their impact on social and political life.
Five key variables representing distinct dimensions of the moral and cultural landscape were selected for the analysis. To ensure intuitive interpretability of the geometric space, specific variables were reverse-coded so that higher numerical values consistently reflect a higher intensity of the measured trait.
The following indicators were included:
To facilitate a longitudinal comparison of cultural drift, the time series data were aggregated into two distinct epochs:
A strict inclusion criterion was applied: only sovereign states possessing valid data points for both time periods were retained. This filtering process resulted in a balanced dataset of 67 countries, ensuring that the observed shifts are attributable to internal cultural evolution rather than sample composition changes.
The analytical framework consists of two sequential steps:
First, Metric Multidimensional Scaling (MDS) was employed to reduce the five-dimensional cultural space into a two-dimensional coordinate system. Prior to dimensionality reduction, all variables were standardized (-score normalization) to eliminate scale differences. Euclidean distance was used as the dissimilarity measure.
Second, Procrustes Analysis was applied to compare the cultural configurations of \(T_1\) and \(T_2\). Since MDS maps are invariant to rotation and reflection, direct coordinate comparison is invalid. Procrustes analysis solves this by finding the optimal translation, rotation, and scaling to superimpose the configuration (2000) onto the configuration (2020). The “Procrustes residuals” (visualized as vectors) represent the true magnitude and direction of cultural change over the two decades.
First, the complex, five-dimensional cultural space was reduced into an interpretable two-dimensional map. The quality of this reduction was evaluated by correlating the original Euclidean distances with the distances preserved in the MDS configuration.
An \(R^2\) value of 0.8304 was obtained. This result indicates that approximately 83% of the variance in the original cultural distances is successfully captured by the two-dimensional model. Consequently, the resulting map is considered a highly reliable representation of the global moral landscape, allowing for valid geometric interpretation.
To quantify the total magnitude of cultural shift between the “Turn of the Century” (2000) and the “Modern Era” (2020), Procrustes statistics were calculated.
## [1] "Procrustes Analysis Result:"
##
## Call:
## procrustes(X = mds_2000, Y = mds_2020, symmetric = FALSE)
##
## Procrustes sum of squares:
## 121.9
A Procrustes Sum of Squares (SS) of 121.89 was recorded across the 67 analyzed nations. This metric represents the minimized sum of squared distances between the aligned configurations. While this value confirms that a measurable shift has occurred, the relatively low Root Mean Squared Error (RMSE = 1.35) suggests that the global “constellation” of cultures has not disintegrated. Instead, the structural relationships between countries (e.g., the relative distance between Scandinavian and Latin American countries) have remained somewhat stable, even as the entire system drifts.
The trajectory of cultural evolution is visualized in the Procrustes plot below. In this diagram:
Three critical observations can be derived from the visual analysis:
The objective of this study was to empirically test the stability of cultural values and to visualize the trajectory of moral change across 67 nations between the turn of the century and the modern era. Based on the application of Multidimensional Scaling and Procrustes Analysis, several key conclusions are drawn.
The hypothesis of static cultural persistence is effectively challenged by the empirical evidence. The graphical analysis confirms that national cultures are not fixed entities, rather, they are subject to a measurable “drift.” The presence of distinct vectors (arrows) for nearly all analyzed countries indicates that moral values, specifically those regarding religion, trust, and civic morality, have evolved significantly over the two-decade period.
A directional consistency was observed in the movement of the majority of nations, lending support to Inglehart’s Modernization Theory. The dominant vector of change suggests a global transition from traditional, survival-oriented values toward secular-rational and self-expression values. However, it is noted that this process is not uniform, while the direction is shared, the velocity of change varies substantially between established post-industrial democracies and developing economies.
Methodologically, the utility of Procrustes Analysis for longitudinal social research has been demonstrated. Unlike static “snapshots” which merely show differences between countries at a single point in time, the Procrustes approach successfully isolated the vector of change. By minimizing the sum of squared differences (SS=121.89), the analysis filtered out statistical noise, allowing for a precise visualization of the true cultural shifts.