A new measure of skill mismatch: theory and evidence from PIAAC

Table 7 Alternative estimates of the skill requirements

	\(\widetilde {min}_{j}-\widehat {min}_{j}\) ^a	\(\frac {\widetilde {min}_{j}-\widehat {min}_{j}}{\widehat {min}_{j}} \)	\(\widetilde {max}_{j}-\widehat {max}_{j}\) ^b	\(\frac {\widetilde {max}_{j}-\widehat {max}_{j}}{\widehat {max}_{j}} \)
Literacy	126.04	0.669	−116.36	−0.365
	(1.533)	(0.013)	(1.235)	(0.004)
Numeracy	140.95	0.852	−111.74	−0.350
	(1.608)	(0.036)	(1.349)	(0.004)

Bootstrapped standard errors in parentheses. See Appendix for details on the bootstrap procedure. All figures are averages over occupational categories and countries. Source: OECD Survey of Adult Skills (PIAAC)
^a \(\widehat {min}_{j}\) = the 5th percentile of the proficiency distribution of workers not feeling able to do more demanding jobs nor feeling the need of training (by occupation). \(\widetilde {min}_{j}\) = the 95th percentile of the proficiency distribution of workers feeling the need of further training
^b \(\widehat {max}_{j}\) = the 95th percentile of the proficiency distribution of workers not feeling able to do more demanding jobs nor feeling the need of training. \(\widetilde {max}_{j}\) = the 5th percentile of the proficiency distribution of workers feeling able to do more demanding jobs