Fig 1.
This figure is a compilation of past findings by the author.
Fig 2.
Upward mobility around the world.
Notes: The data for this estimation comes from the pre-modelled dataset (see S2 Appendix). The method used for comparing means is ANOVA and p is its p-value.
Fig 3.
How do causal machine-learning techniques address confounding problems?.
Notes: In the given figure, the primary effect of interest (denoted as 1) is the impact of D on the outcome variable Y. However, this effect is confounded by variables X that influence both D and Y (denoted as 2 and 3). To address this problem, causal machine learning is utilized, fundamentally consisting of two parts. The machine learning component used to generate D’ and Y’–these are the corresponding parts of D and Y that are explained by the confounders X, and the parameters of this process are referred to as the nuisance parameters (such as the indices representing the impact of X on D and Y). The causal inference component will perform estimations from the residuals of Y and D after removing D’ and Y’. The orthogonalization technique used in the process ensures that the target effect is not influenced by confounders (in other words, it is invariant to the nuisance parameters). It is also noted that these causal machine-learning techniques only report the target effect.
Fig 4.
What really matters for intergenerational mobility?.
Notes: The sample size N = 6,725. The data for this estimation is derived from the pre-modelled dataset (refer to S2 Appendix). R represents Pearson’s correlation coefficient, and p denotes its p-value.
Fig 5.
Does education expansion work for promoting upward mobility?.
Notes: This Fig refers to the table format provided in S4 Appendix. The sample size N = 6,725. In this study, the outcome variable is upward mobility. For each estimate, the treatment variables (target variables) are either (education) inequality, (education) expansion, or (parental) dependency. The confounding variables consist of all other variables present in the dataset, which are not directly considered treatment or outcome variables.
Fig 6.
Gender bias in upward mobility.
Notes: The sample size N = 6,725. The data for this estimation comes from the pre-modelled dataset. The method used for comparing means is ANOVA and p is its p-value.
Fig 7.
Is intergenerational mobility still gender-biased?.
Notes: refer to the table format provided in S5 Appendix. The sample size N = 6725. The outcome variable is upward mobility.
Fig 8.
The multidimensional gender bias in intergenerational mobility.
The data source for these charts is GDIM 2023. The chart on the left computes intergenerational mobility using the education levels of daughters. Conversely, the chart on the right uses the education levels of sons and all children in general to compute intergenerational mobility.