The author provides a review of ten adult attachment measures from the perspective of psychometric properties and theoretical grounding. The measures examined are:

Apart from considering each measure individually, the author makes the general points that there are serious limitations associated with the use of categorical measures (AAQ, RQ) and that some measures can be difficult to interpret (AAS, IPPA) as they produce results that do not easily map onto existing conceptualisations of attachment.

Considerations when Choosing an Instrument:

• Instruments should be chosen that reflect the theoretical stance of the project. Some instruments focus on measuring attachment behaviours (directed towards mainting proximity) and others are more concerned with attachment style (experience on separation, loss and reunion).
• It should be noted that some measures focus on the parental relationship (PBI, AAI, IPPA, PAQ), other on romantic pair-bond and other close, adult relationships (AAQ, AAS, MIA, RQ, AI), whilst some focus on more general relationships (BORI).
• Some measures require retrospective accounts (PBI, AAI) whilst others tap perception of current relationships (IPPA, PAQ, AAQ, AAS, MIA, RQ, AI, BORI). It should be noted that these are subject to different confounding factors. For example, mood state has been found to effect recollections of past relationships (Kenny & Rice, 1995).
• Other questions that must be addressed concern the nature of the theoretical construct being tapped. For example, studies with high face validity can be assumed to be measuring conscious attitudes, whereas Bartholomew (1994) has questioned the extent to which working models of attachment are consciously available.

Overall, Garbarino concludes that great care should be taken in choosing attachment measures for a particular study. As a general guide for clinical practice, she recommends:

• BORI and MIA for assessing relationship between personality (particularly pathological personality patterns) and relationship functioning.
• The AI or the AAI for comparing attachment with attitudes towards love and relationships.
• The MIA and AI for assessing relationship behaviours in a clinical setting.
• The IPPA, PAQ and PBI for assessing relationships within families.

Note that Garbarino refers to Hazan and Shaver's (1987) categorical attachment measure as the 'Attachment Style Measure (ASM)'. To avoid confusion with Simpson's (1990) measure of the same name, it had been referred to here by its alternative title of Adult Attachment Questionnaire (AAQ).