AI-powered recruitment is linked to positive organizational shifts and improved career outcomes in Nigerian universities; the effects are strongest when employees have higher emotional intelligence and stronger family support.

Main Finding

Adopting AI in recruitment (AIREC) in Nigerian higher-education institutions is strongly associated with “positive organizational shock” (POSH), and both direct and indirect pathways link AIREC to improved perceived career development (CD). POSH partially mediates the AIREC→CD relationship (VAF = 57.31%). Emotional intelligence (EI) strengthens the AIREC→POSH effect; family support (FS) strengthens the POSH→CD effect.

Key Points

• Strong direct effects:
  • AIREC → POSH: path = 0.744, t = 14.836, p < .001.
  • POSH → CD: path = 0.453, t = 5.535, p < .001.
  • AIREC → CD (direct): path = 0.251, t = 3.216, p = .001.
• Mediation:
  • Indirect AIREC → POSH → CD: path = 0.337, t = 5.065, p < .001.
  • Variance accounted for (VAF) = 57.31% → partial mediation by POSH.
• Moderation:
  • EI × AIREC → POSH: path = 0.056, t = 2.073, p = .038 (EI amplifies AIREC’s positive organizational effects).
  • FS × POSH → CD: path = 0.103, t = 2.566, p = .01 (family support amplifies translation of POSH into career gains).
• Model explanatory power:
  • R²(CD) = 0.782 (substantial).
  • R²(POSH) = 0.664 (moderate).
• Measurement and fit:
  • Reliability: Cronbach’s α and composite reliability > 0.7; AVE > 0.5; outer loadings > 0.6; VIF < 5.
  • Fit: SRMR = 0.072 (saturated), 0.082 (estimated); chi-square and NFI reported as supportive.

Data & Methods

• Context: Higher-education institutions in Nigeria; study framed post-COVID-19.
• Sample: N = 400 human resources practitioners across 8 universities, proportionate sampling based on university populations.
• Design: Cross-sectional survey using 5-point Likert items.
• Constructs & measures:
  • AIREC: use of AI for screening, scheduling, predictive analytics (custom items).
  • POSH: unexpected positive organizational changes (adapted from Seibert et al., 2013).
  • CD: perceived career advancement, skills development, job satisfaction.
  • EI: measured via Law et al. (2004) dimensions.
  • Family Support: emotional/financial/practical support (King et al., 1995).
• Analysis: Partial Least Squares Structural Equation Modeling (PLS-SEM) with bootstrapping to estimate paths, moderation, mediation, and model fit.

Implications for AI Economics

• Micro/labor-market implications:
  • AI-driven recruitment can generate organizational benefits that translate into perceived career gains for incumbent staff—suggests productivity and human-capital effects beyond simple hiring efficiency.
  • EI and family support as moderators imply distributional heterogeneity: personal and household resources affect how workers capture gains from AI adoption.
• Policy and management:
  • Complementary human-capital investments (EI training, family-support programs, SHRM practices) can increase the returns to AI adoption and reduce unequal uptake of benefits.
  • Implementation should pair technological rollout with fairness, transparency, and worker-support measures to maximize positive organizational shocks.
• Research and evaluation needs for AI economics:
  • Move beyond self-reported perceptions toward objective outcomes: hires, promotions, wages, retention, and productivity measures.
  • Longitudinal or experimental designs to establish causal timing (e.g., whether AI adoption drives POSH or vice versa).
  • Heterogeneity analysis: gender, rank, discipline, and socioeconomic status to assess inequality risks from AI-enabled hiring.
  • Cost–benefit and general-equilibrium analyses: how AI recruitment affects labor market matching, search frictions, and aggregate human-capital accumulation.
• Limitations to consider when applying results:
  • Cross-sectional and self-report data from HR practitioners in Nigerian universities limit external validity (other sectors/countries may differ).
  • Possible social desirability or common-method bias; mechanisms of algorithmic fairness and bias were discussed but not empirically measured.
  • Magnitudes: moderation effects, while statistically significant, are small (EI moderation path = 0.056), so practical significance should be tested in follow-ups.

Overall, the study offers evidence that AI recruitment can produce organizational shocks that help career development, but realizing those gains depends on human-capital and family-context complements. For AI economics, this highlights the importance of complementarities and distributional channels when evaluating the labor-market effects of AI adoption.