Background There is a degree of affective interdependence that is considered normal and only becomes pathological if it causes excessive suffering, both for the subject and for those close to them. Our objective was to introduce and psychometrically validate a short and effective affective dependency scale, the Affective Dependence Scale (ADS-9). Methods We used a sample of 762 participants (clinical: emotional dependent subjects n = 212, comparison: non-emotionally-dependent addicted subjects n = 272, and general population n = 278) to assess the factor structure, the psychological construct validity and the measurement invariance for the ADS-9 by means of independent exploratory factor analyses for each sample group and subsequent multigroup confirmatory factor analyses. Results Our results confirm that ADS-9 is a psychometrically consistent instrument, with construct and clinical validity, as well as configural, metric and scalar invariance across different sample groups (clinical, comparison and general population). A hypothesized two-dimensional structure was confirmed by means of factor analyses. Both sub-scales of this abbreviated form, Submission and Craving, showed a good agreement with the previously validated Relationships and Sentimental Dependencies Inventory (IRIDS-100). Conclusion The ADS-9 is a brief instrument that appears to reliably detect the dependent and pathological components of affective dependence. It consists of two sub-scales, describing Submission (adaptation, accommodation, and subjugation) and Craving (imperative need for the other with the presence of disturbing states). We suggest that it is a versatile scale that may be useful for clinicians and researchers.