Maternal psychopathology as related to child-mother quality of attachment in a preschool sample

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)


Counseling Psychology

First Committee Member

Carolyn S. Garwood - Committee Chair


The relation between maternal personality and child quality of attachment was investigated in a sample of preschoolers and their mothers. Based on studies of infants, it was expected that approximately two thirds of children of mothers with no evidence of psychopathology would be securely attached. Based on studies of preschool children of depressed mothers, it was anticipated that approximately one half to two thirds of children whose mothers evidenced psychopathology would be anxiously attached. It was also expected that those mothers evidencing more chronic and severe psychopathology would have children who were more likely to display controlling-disorganized or anxious/combination attachments.Participating families were selected from a larger research study entitled "The Family Project". The families represented a wide range of income and ethnic groups. In addition, a small clinical subsample of maltreating families and families with a child in mental health treatment was included.Child quality of attachment to the mother was assessed by means of the MacArthur three-year Strange Situation procedure. This assessment is presently undergoing validation in this and other studies. Maternal personality was assessed using the Millon Clinical Multiaxial Inventory (MCMI), a self-report personality inventory corresponding to the diagnostic criteria of the DSM-III.To test the first hypothesis, maternal psychopathology scores on the MCMI were classified as normal or pathological and child security of attachment as secure or insecure. No significant relation was found between personality pathology and the security of attachment.To test the second hypothesis, specific personality scales on the MCMI were compared to MacArthur A/B/C/D groupings. No relation was found between chronicity or severity of personality disorder and pattern of attachment.Post hoc analyses explored three possible explanations for the unexpected findings. Specifically, in order of their significance: (1) The MacArthur system appears to be inappropriate for use with a heterogeneous sample; (2) The variables were indirectly and distally related and required intermediary variables to explain their relation; (3) A larger, clinically diverse sample was needed for comparisons between specific MCMI scales and attachment classifications.Among the explanations considered here, the lack of validity of the MacArthur three-year Strange Situation with a heterogeneous sample seems to be the primary reason for failure to confirm the hypotheses.


Education, Educational Psychology; Psychology, Clinical

Link to Full Text


Link to Full Text