APPLICANT TRACKING SYSTEM
Department
Year of Publication
Keyword
Publication Type
Abstract
Applicant Tracking Systems (ATS) have become integral to modern recruitment processes, yet
their tendency to systematically reject qualified candidates through over-filtering remains a
critical concern. This study investigated the prevalence, causes, and impact of false rejection in ATS using a dataset of 600 organizations, of which 443 had adopted ATS technology. Statistical analysis using IBM SPSS Statistics (Version 29) revealed that over-filtering affects 70.4% of ATS users, compromising candidate selection in approximately 52% of all organizations
surveyed. The analysis identified cultural bias in language processing (11.9%) and educational
background over-filtering (10.6%) as the leading causes of false rejection, followed by industry- specific terminology barriers (6.8%) and resume parsing formatting issues (6.5%). Notably, while 100% of organizations experiencing bias implemented mitigation measures, none reported
successful elimination of bias, indicating that over-filtering represents a persistent structural
feature rather than a correctable implementation flaw. Comparative severity analysis revealed
that high-intensity biases requiring algorithmic reengineering, particularly cultural and
demographic biases demonstrated the strongest resistance to remediation despite intensive
intervention. The study concludes that achieving fair candidate selection requires fundamental
redesign of ATS architecture with fairness as a core principle, moving beyond procedural
adjustments toward culturally competent Natural Language Processing engines, transparent
auditable algorithms, and meaningful human oversight at critical screening stages. Keywords: Applicant Tracking Systems, false rejection, over-filtering, algorithmic bias, recruitment technology, candidate screening, cultural bias, resume parsing
their tendency to systematically reject qualified candidates through over-filtering remains a
critical concern. This study investigated the prevalence, causes, and impact of false rejection in ATS using a dataset of 600 organizations, of which 443 had adopted ATS technology. Statistical analysis using IBM SPSS Statistics (Version 29) revealed that over-filtering affects 70.4% of ATS users, compromising candidate selection in approximately 52% of all organizations
surveyed. The analysis identified cultural bias in language processing (11.9%) and educational
background over-filtering (10.6%) as the leading causes of false rejection, followed by industry- specific terminology barriers (6.8%) and resume parsing formatting issues (6.5%). Notably, while 100% of organizations experiencing bias implemented mitigation measures, none reported
successful elimination of bias, indicating that over-filtering represents a persistent structural
feature rather than a correctable implementation flaw. Comparative severity analysis revealed
that high-intensity biases requiring algorithmic reengineering, particularly cultural and
demographic biases demonstrated the strongest resistance to remediation despite intensive
intervention. The study concludes that achieving fair candidate selection requires fundamental
redesign of ATS architecture with fairness as a core principle, moving beyond procedural
adjustments toward culturally competent Natural Language Processing engines, transparent
auditable algorithms, and meaningful human oversight at critical screening stages. Keywords: Applicant Tracking Systems, false rejection, over-filtering, algorithmic bias, recruitment technology, candidate screening, cultural bias, resume parsing
Supervisor(s)
co-supervisor