Methods
Summary
The study will be conducted entirely online using two recruitment strategies running simultaneously. Participants will be recruited through CloudResearch, a validated online research platform commonly used in academic studies, as well as through social media platforms including Facebook, Reddit, and LinkedIn. These platforms were chosen for the diversity of individuals on each platform.
All participants will complete the survey through SurveyMonkey, which provides secure, anonymous data collection. The survey takes approximately 20-25 minutes. It includes the proposed CENOS items, along with three established measures: the Parental Bonding Instrument (PBI), the Guilt and Shame Proneness Scale (GASP), and the Toronto Alexithymia Scale-20 (TAS-20). These established measures allow us to evaluate whether the CENOS is measuring what it's designed to measure.
The target sample is approximately 200-300 adults aged 18 and older in the United States who can read and understand English. Participation is completely voluntary and anonymous.
Data will be analyzed using Jamovi, a free, open-source statistical software package, and will include exploratory factor analysis to examine the scale's structure, Cronbach's alpha to evaluate internal consistency, and mediation analysis using bootstrapping procedures to test the proposed theoretical pathway.
Challenges
The primary anticipated challenges of this pilot study are inherent to the nature of retrospective self-report methodology and early-stage measure development. Retrospective recall bias represents a meaningful limitation; participants are asked to reflect on childhood caregiver behaviors, which may be influenced by current mood states or the passage of time. This is partially mitigated by the behaviorally specific design of CENOS items, rather than by subjective emotional recollections, thereby reducing reliance on interpretive memory.
Self-selection bias is a second anticipated challenge, as individuals who choose to participate in a study about childhood experiences and adult relationships may disproportionately represent those with more pronounced histories of emotional neglect. While this cannot be fully eliminated in convenience sampling, the dual recruitment strategy combining CloudResearch's verified participant pool with social media recruitment is designed to maximize sample diversity.
Finally, as a first-generation measure, the CENOS items may require refinement following pilot data collection. This is an expected outcome of pilot research; the purpose of this phase is to identify areas for improvement before larger-scale validation studies. Findings will directly inform item revision and future research phases.
Pre Analysis Plan
The pilot study will employ several complementary analytical approaches to evaluate the psychometric properties of the CENOS and provide preliminary evidence for the Omission of Relational Repair model of CEN.
Preliminary analyses will include examination of missing data patterns, distributional properties, and descriptive statistics for all study variables to ensure data quality prior to primary analyses.
Internal consistency will be evaluated using Cronbach's alpha coefficients for the total CENOS scale and each subscale independently — the Repair Omission Index (ROI) and the Relational Repair Index (RRI) — with acceptable reliability defined as α ≥ 0.70.
Exploratory Factor Analysis (EFA) will be conducted to examine the underlying structure of the CENOS and determine whether items load onto the proposed domains. Given the pilot nature of this study, EFA is appropriate as the factor structure has not yet been empirically confirmed.
Convergent validity will be assessed through correlations between CENOS scores and the established measures included in the survey — the PBI, GASP, and TAS-20. Significant correlations in the expected directions will provide preliminary evidence that the CENOS is measuring theoretically related constructs.
Mediation analysis using bootstrapping procedures will be conducted to test the core hypothesis of the ORR model, that maladaptive shame-driven response patterns mediate the relationship between childhood emotional neglect and impaired adult relational repair capacity.
All analyses will be conducted using jamovi statistical software. Given the pilot nature of this study, findings will be interpreted with appropriate caution and used primarily to inform refinement of the CENOS for future validation research.
