NYC Local Law 144 Bias Audit Results
This page contains the bias audit results for automated employment decision tools (AEDTs) used by Meridian Talent Solutions, as required by NYC Local Law 144.
Selection Rates by Protected Category
The following table shows selection rates for different demographic categories, calculated in accordance with EEOC guidelines.
By Sex
| Category | Total Applicants | Selected | Selection Rate | Impact Ratio |
|---|---|---|---|---|
| Male | 1,847 | 312 | 16.89% | 1.00 |
| Female | 1,623 | 259 | 15.96% | 0.94 |
By Race / Ethnicity
| Category | Total Applicants | Selected | Selection Rate | Impact Ratio |
|---|---|---|---|---|
| White | 1,102 | 198 | 17.97% | 1.00 |
| Black or African American | 724 | 118 | 16.30% | 0.91 |
| Hispanic or Latino | 689 | 108 | 15.67% | 0.87 |
| Asian | 612 | 103 | 16.83% | 0.94 |
| Native Hawaiian or Pacific Islander | 87 | 14 | 16.09% | 0.90 |
| American Indian or Alaska Native | 64 | 10 | 15.63% | 0.87 |
| Two or More Races | 192 | 20 | 10.42% | 0.58 |
Impact Ratio: Calculated as the selection rate for each category divided by the selection rate of the highest-performing category. Under the EEOC "4/5ths rule," an impact ratio below 0.80 may indicate adverse impact requiring further investigation. Categories with fewer than 5 applicants are suppressed per statistical best practices.
Note: The "Two or More Races" category shows an impact ratio of 0.58, which is below the 0.80 threshold. This finding has been documented in the full audit report with a remediation plan. The employer has initiated a review of the AEDT scoring algorithm for this category.
Audit Methodology
Bias audit conducted using HAIEC deterministic analysis engine (v2025.1.0). Selection rates calculated per EEOC Uniform Guidelines on Employee Selection Procedures. Impact ratios computed using the 4/5ths rule. Statistical significance assessed using Chi-Square test (n ≥ 30) and Fisher's Exact test (n < 30) at p < 0.05.
Historical applicant data from January 1, 2025 through December 31, 2025. Data extracted from employer's applicant tracking system (ATS) via structured CSV export. 3,470 total applicant records across 3 job categories: Software Engineering, Product Management, and Customer Success.
Chi-Square test with Yates' correction applied to all categories with expected cell frequencies ≥ 5. Fisher's Exact test applied to categories with small sample sizes. Bonferroni correction applied for multiple comparisons. All calculations are deterministic and reproducible (SHA-256 verified).
Independent Auditor
AI Compliance Officer, HAIEC
Independent auditor with no financial relationship to the AEDT developer. Audit conducted using HAIEC Deterministic Analysis Engine v2025.1.0.
Certified: This bias audit was conducted by HAIEC's AI Compliance Officer serving as the independent auditor under NYC Local Law 144. The auditor maintains independence from the AEDT developer and has personally reviewed and certified all findings in this disclosure. Audit date: February 1, 2026.