“…One is that we did not have attachment classification measures at the early childhood assessment, and so are unable to contrast stability in caregiving with stability in attachment classifications. Second, although attachment research is largely influenced by interview and narrative assessments (Allen, Moore, Kuperminc, & Bell, ; Joseph, O'Connor, Briskman, Maughan, & Scott, ; Kobak & Sceery, ), other kinds of measures exist (Kim, Boldt, & Kochanska, ) and need to be incorporated in longitudinal research linking caregiving and attachment across multiple periods in the life course. Additionally, our assessment of caregiving did not incorporate all of the many aspects that may be related to attachment quality (Allen et al., ; Dozier, Albus, Fisher, & Sepulveda, ; Karavasilis et al., ); our reliance on a semistructured play and tidy‐up task may not have optimized the early assessment of caregiver sensitivity.…”