Table 3.

Beta regression results – testing the education-FLFP relationship in Sri Lanka (2017–2024)

VariableModel 1 (Linear)Model 2 (Quadratic)Model 3 (Temporal)Model 4 (Full model)
Constant−0.197 (0.199)−0.158* (0.066)−0.183 (0.196)−0.143*** (0.036)
Education (linear)4.964*** (0.482)4.961*** (0.170)4.485*** (1.343)4.691*** (0.163)
Education (quadratic)2.826*** (0.148)2.823*** (0.083)
Education (cubic)1.003*** (0.083)
Year fixed effectsYesYesYesYes
Education × yearYesYes
Pseudo R²0.7350.960.7360.992
Log-likelihood36.35275.48236.454105.437
AIC−52.704−128.963−38.908−172.874
BIC−35.815−110.385−10.197−140.785
N40404040
Note(s):

p-values are based on wild cluster bootstrap inference at the education-category level (Five clusters; B = 999), which is appropriate for few-cluster settings. Standard errors (in parentheses) are cluster-robust at the education level. All models use orthogonal polynomial specifications for education and include year fixed effects. ***p < 0.001, **p < 0.01, *p < 0.05

Source(s): Prepared by authors

or Create an Account

Close subscription notice
Close access options