This site is a work in progress and has not been widely shared. Content may contain errors. Feedback is welcome.
This site is undergoing review. Some annotations were human-generated, some AI-generated — all are being verified.
Back to papers

The Demand for AI Skills in the Labor Market

Alekseeva, Azar, Gine, Samila, Taska

2019IZA Discussion Paper Series24 citations
Observational labor market
GenderJunior / entry-level
DOI: 10.2139/ssrn.3470610
Summary

Contini et al. use multinomial logit models on Italian administrative education data linking student performance records with high school enrollment to study how mathematics and language skills explain gender differences in the choice of STEM versus humanities tracks at age 14.

Main Finding

Girls require significantly stronger signals of mathematical ability than boys to choose STEM high schools, with school performance explaining only a small portion of the gender gap overall, though it explains more for children of highly educated parents (16-30% of gap for Traditional STEM Lyceum) than for disadvantaged backgrounds.

Primary Datasets

Anagrafe Nazionale Studenti (Italian National Register of Students); INVALSI standardized test scores

Secondary Datasets

None

Key Methods
Multinomial logit models with performance measures (teacher grades and standardized test scores) as predictors of high school track choice; Oaxaca-Blinder decomposition for categorical outcomes; analysis stratified by parental education
Sample Period
2013-2017
Geographic Coverage
Italy (Piedmont, Lombardy, Veneto regions)
Sample Size
168,445 students in grade 8 (2015-16 school year) across 1,837 middle schools
Level of Analysis
Individual
Occupation Classification
None
Industry Classification
None
Notes
Labour Economics [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects. [Claude classification]: CRITICAL DATA QUALITY ISSUE: This paper has NO connection to AI whatsoever. It studies gender gaps in educational track choices (STEM vs humanities) among Italian 14-year-olds. The existing metadata references a completely different paper (Alekseeva et al. on AI skills demand). This appears to be a severe mismatch between paper metadata and actual PDF content. The paper should NOT be in an 'Empirical Economics of AI' dataset. Classification provided reflects actual paper content on educational gender gaps, not AI labor market effects.