Harmonizing Discordant Notes: A Reply to Waters

THoMPSON, Ross A.; LAMB, MICHAEL E.; and EsTEs, DAvm. Harmonizing Discordant Notes: A Reply to Waters. CHILD DEVELOPMENT, 1983, 54, 521-524. The major concerns raised in Waters's commentary on our earlier report on attachment stability are discussed and shown to be incorrect. First, studies showing high stability in attachment classifications have employed a range of sampling and scoring procedures that may help explain why they obtained high stability estimates. Thus, attachment stability in the secondyear is not a well-replicated finding. Second, changes in attachment status were systematically related to changing family circumstances that directly affect mother-infant interaction, such as maternal employment and nonmaternal care. Bidirectional changes in attachment status are reasonable considering the range of possible reactions to these events in middle-class families. Events occurring prior to the first assessment may also have affected stability by introducing short-term, transient changes in the security of attachment. Third, measurement error is unlikely to be a problem with our study, since there is good reason for confidence in the accuracy of our scoring procedure. In sum, our findings affirm the importance of viewing mother-infant attachment as a dynamic relationship, responsive to family conditions, and these findings are consistent with those of other researchers.