Background Lack of adequate parental care is a consistent predictor of adult depression. Questionnaire measures that compare well with interviews are needed for large-scale studies of affective disorders in the general population. A widely used questionnaire, the Parental Bonding Instrument (PBI), was compared with a detailed standardized interview, the Childhood Experiences of Care and Abuse (CECA). Methods PBI ratings of maternal and paternal care were obtained from 192 women aged 25–36, identified from primary care lists. The women were interviewed for childhood neglect using the CECA, and ratings made blind to their PBI responses. Results The discriminative ability of PBI care scores to predict measures of neglect in the CECA were moderate to high, and the addition of paternal scores did not add to the prediction from maternal scores. Shortened forms of the PBI maternal care scales provided comparable predictions to those from the full scale, particularly three items from the maternal care scale, identified by logistic regression. Limitations PBI and CECA measures on the women were retrospective. Low numbers in some of the subgroups of interest limited statistical power and is reflected in wider confidence intervals. The three maternal care items identified by logistic regression need to be confirmed in other samples as being as efficient as the combined maternal and paternal care scale scores. Conclusions The maternal care scale of the PBI compares reasonably well as an index of overall neglect in childhood to that provided by the CECA.