Methodology
How each instrument is scored, how reports are generated, and what the results mean — and don't mean.
Instrument Scoring
Each instrument uses its published scoring algorithm. Scores are not adjusted, curved, or normalized in ways that deviate from the original research design. What follows is a brief description of each instrument's scoring approach.
Big Five (IPIP-NEO)
Five subscale scores (Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism), each based on 12 items. Some items are reverse-scored per the IPIP key. Scores expressed as percentile within general population norms.
Goldberg, L. R. (1999). A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several five-factor models. Personality Psychology in Europe, 7, 7–28.
Attachment Style (ECR-RS)
Anxiety and avoidance scores computed per relationship domain (romantic, family, friends, peers). Domain-level scoring allows for relationship-specific attachment profiles rather than a single global attachment style.
Fraley, R. C., Heffernan, M. E., Vicary, A. M., & Brumbaugh, C. C. (2011). The Experiences in Close Relationships–Relationship Structures questionnaire. Psychological Assessment, 23(3), 615–625.
Personal Values (PVQ-40)
Ten value subscores. Within-person centering is applied: each score is adjusted relative to the respondent's own mean, to control for acquiescence bias and allow meaningful cross-value comparison within the individual profile.
Schwartz, S. H. (2012). An Overview of the Schwartz Theory of Basic Values. Online Readings in Psychology and Culture, 2(1).
Conflict Style (TKI)
Five mode scores (competing, collaborating, compromising, avoiding, accommodating) based on item selection frequencies. Minimum 10 responses required for a reportable result; insufficient data flagged rather than estimated.
Thomas, K. W., & Kilmann, R. H. (1974). Thomas-Kilmann Conflict Mode Instrument. Xicom.
Love Languages
Five language scores based on selection frequency. Note: while widely used, the Love Languages framework has more limited peer-reviewed psychometric validation than instruments such as IPIP-NEO or ECR-RS. Results describe preferences, not diagnoses.
Chapman, G. (1992). The Five Love Languages. Northfield Publishing. Psychometric development: Egbert, N., & Polk, D. (2006). Speaking the language of relational maintenance.
Enneagram
Nine type scores; highest score identifies primary type. Wing types scored from adjacent types. Tritype derived from highest-scoring type in each of three centers (instinctive, feeling, thinking). Note: the Enneagram has limited peer-reviewed clinical validation. Results describe motivational patterns, not clinical diagnoses.
Riso, D. R., & Hudson, R. (1996). Personality Types. Houghton Mifflin. Psychometric validation: Sutton, A., et al. (2020). Personality and the Enneagram. Journal of Individual Differences.
Communication Style
Four style scores (expressive, driver, analytical, amiable). Primary style identified by highest score. Scores reflect behavioral tendencies, not fixed traits — communication style is context-dependent.
Merrill, D. W., & Reid, R. H. (1981). Personal Styles and Effective Performance. CRC Press.
Holland Codes (RIASEC)
Six interest area scores (Realistic, Investigative, Artistic, Social, Enterprising, Conventional). Top two or three codes form the RIASEC profile used for career matching.
Holland, J. L. (1997). Making Vocational Choices: A Theory of Vocational Personalities and Work Environments (3rd ed.). Psychological Assessment Resources.
Report Generation
After scoring, your instrument results are assembled into a profile object that contains your scores, subscores, and data quality indicators (completion rate per instrument, confidence levels per scale). This profile is passed to Claude (Anthropic's AI model) with explicit instructions to anchor every insight to your actual score data.
Reports are generated in twelve sections covering personality traits, relationship patterns, values, conflict style, communication, career interests, and cross-instrument interactions. Each section is validated against a schema before being stored — the model cannot output generic text unanchored to your data.
AI-generated content is a synthesis tool. It organizes and interprets your scores. It does not add information that isn't in your responses. Where your data is ambiguous or incomplete, the report notes this rather than filling in gaps with assumptions.
Limitations
Self-report bias. All instruments rely on how you describe yourself. Your scores reflect your self-perception, which may differ from how others perceive you or from behavioral observation.
State vs. trait. Some traits (especially Neuroticism and Extraversion in Big Five) can shift during periods of high stress, illness, or major life change. Scores represent your current self-perception, not a fixed lifetime trait score.
Instrument validation variation. The Big Five, ECR-RS, PVQ-40, TKI, Communication Styles, and Holland Codes have extensive peer-reviewed validity evidence. The Enneagram and Love Languages are widely used but have more limited clinical validation. We report both; we do not conflate them.
Not a clinical tool. Your True Self reports are for self-understanding. They are not psychological evaluations, clinical diagnoses, or medical advice. They are not appropriate for hiring, legal, custody, or clinical decision-making. If you are experiencing mental health difficulties, please consult a licensed professional.
Editorial Process
Guide articles on this site are researched and written with reference to the primary literature for each instrument. Where specific claims are made about instrument validity, reliability, or research findings, citations are provided in the article text. Guide articles are periodically updated as new research becomes available.
Questions about methodology or instrument selection: team@yourtrueself.app