Screening & Vetting

Behavioral Analysis of Background Investigations (BABI) Study Phases I and II

The National Security Background Investigations and Adjudications process is the entry point through which individuals are vetted to obtain eligibility to access classified government information. Established after World War II and refined over time by court decisions, executive orders, and agency directives, national background investigations examine personal behaviors, relationships, and activities that may increase risk to safeguarding classified information. Background investigation activities include review of SF-86 form, interview with applicants, interviews with references, and records checks to validate personal information and identify undisclosed risks. Over time, societal changes have increased both the volume and time required to collect case information. Combined with the Office of Personnel Management data breach in 2014 and the subsequent backlog of more than 700,000 background investigation cases, these pressures highlighted the need to better understand the personnel vetting process and data being collected. MITRE’s Insider Threat Research & Solutions team conducted the Behavioral Analysis of Background Investigation (BABI) study from 2022-2025 to identify behaviors, activities, events, and associations most relevant to adjudicative outcomes, with the goal of reducing the time and effort required to process background investigations. The Department of War (DOW) provided MITRE data from more than 47,000 background investigation cases. Discussions with DOW and initial exploration of the case data indicated that the background investigation (BI) data were not consistently structured and, therefore, had not been systematically analyzed to identify key investigative data elements associated with adjudicative outcomes. The team of behavioral scientists and data scientists proceeded to standardize and structure the application data (e.g., SF-86 responses), interview notes, and records data. The resulting structured data were manually annotated using rigorous qualitative methods and then analyzed to identify differences in key factors between favorable and derogatory adjudicated case outcomes. Findings and recommendations for Phase I were presented to senior leaders at the Office of the Under Secretary for War Intelligence & Security (OUSW(I&S)) and the Defense Counterintelligence and Security Agency (DCSA). This phase resulted in the current BABI Dataset.

In Phase 2, the BABI study focused on detailed analysis of key questions about the background investigation process as identified by government stakeholders. This phase consisted of a series of short analytical sprints that addressed the following topics:

Comparative Analysis of Virtual and In-Person Interview Protocols
The comparative analysis was conducted to understand from a data perspective quality and quantity differences between BI information received during virtual and in-person investigator interviews with the applicant and references.
Exploratory Analysis of Foreign Nexus Data
Potential foreign influence associations and foreign preference activities continue to concern national security stakeholders, especially as espionage cases continue to come to light. Analysis examined the impact of foreign nexus behaviors, activities, events, and associations provided during the BI process on adjudicative outcomes.
Examination of Cannabis Use Data
The legalization of cannabis, particularly marijuana, by several states has created a complex legal and regulatory landscape, as its use and sale remain prohibited under federal law. This disparity has introduced challenges for security background investigations, necessitating a nuanced approach to evaluating cannabis use information. In response to this complexity, BI stakeholders expressed interest in analyzing cannabis use data within the BABI Dataset to better understand its implications and inform decision-making.
Detailed Analysis of Financial Issues
Stakeholders reported that financial considerations are often cited as an adjudicative guideline for both favorable and derogatory BI cases. A deep-dive analysis of all the financial data identified in the BABI Dataset was conducted to better understand the impact specific types of finances and debt ratings have on adjudicative outcomes.
Detailed Analysis of Criminal Conduct
A deep-dive analysis of all criminal conduct data (including type, frequency, and consequences) identified in the BABI Dataset was performed to better understand the impact specific types of criminal conduct have on adjudicative outcomes.
Expandable Focused Investigation (EFI) Model Flags Identification and Investigative Follow-Up
EFI analysis sought to address stakeholder concerns regarding the traceability and resolution of EFI model flags identified during the BI process. Analyses focused on connections between EFI triggered SF-86 flags and text from records and candidate and reference interviews indicating follow-up during the investigation.

Presentations of the BABI Study for Phase I and for each of the analyses in BABI Phase II can be requested by government personnel security screening and vetting offices.

For more information, please contact Dr. Deanna D. Caputo (dcaputo@mitre.org)

Back to previous page