Retrospective methods for assessing maternal and child dietary diversity and women's time-use are labour-intensive and prone to misreporting. This thesis comprises the first-ever validation of an automated wearable camera (AWC), an inexpensive technology for prospectively and unobtrusively collecting data, in a low-income country (LIC). The cross-sectional study was conducted in rural Uganda with mothers of a child aged 12-23 months (n=211) to (1) assess the feasibility of an AWC-based image-assisted recall (AWC-IAR); and (2) evaluated the concurrent criterion validity of the AWC-IAR method and the 24-hour recall (24HR) method, compared to observation for (a) assessing maternal and child dietary diversity, and (b) women's time use. Agreement was assessed using the Bland–Altman limits of agreement (LOA) approach. Most participants rated the AWC-IAR method as good or very good. Eight participants withdrew from the study and 27 were excluded due to AWC inoperability. The relative bias was low, for maternal and child dietary diversity estimated using semi-quantitative multiple-pass 24HR and image-assisted recall (IAR) methods, but the LOA were wide. For time use, the systematic bias varied from 1 minute (domestic chores) to 226 minutes (caregiving) for 24HR and 1 minute (own production) to 109 minutes (socializing) for IAR. The LOA were within 2 hours for employment, own production, and self-care for 24HR and IAR but exceeded 11 hours (24HR) and 9 hours (IAR) for caregiving and socializing. The AWC-IAR method was feasible in this rural LIC context, although challenges remain. The IAR and 24HR methods provided an accurate estimate of median maternal and child dietary diversity and minutes women spend in employment, own production, and domestic chores. The median time women spent caregiving and socializing was underestimated by both methods. The high LOA indicate that, for research purposes, associations between either dietary diversity or time use and outcomes would be attenuated.