Dot-and-interval chart comparing two denominators for recorded unpaid eldercare in the 2023–2024 American Time Use Survey. The universe is adult providers age 18 or older who had helped at least one person age 65 or older more than once in the preceding three to four months, with a valid diary-care answer. Filled dots average across all days, including days with zero care. Open dots average only diaries with positive recorded care. Lines show 95% confidence intervals using 160 replicate weights. Employed providers with an own resident child under 18: 0.37 hours across all days (95% CI 0.19–0.56), versus 2.44 hours on care days (1.40–3.48); 356 diaries, 53 positive-care diaries, weighted participation 15.4%. Employed providers without an own resident child under 18: 0.71 hours across all days (0.56–0.85), versus 2.94 on care days (2.45–3.42); 1,050 diaries, 265 positive-care diaries, participation 24.0%. Not-employed providers: 1.90 hours across all days (1.63–2.18), versus 4.83 on care days (4.24–5.42); 1,122 diaries, 445 positive-care diaries, participation 39.4%. Six adult providers with an unknown or invalid diary-care answer are excluded. The groups differ in age and other characteristics; no causal effect is estimated. Eldercare can overlap with leisure and other primary activities and does not measure all responsibility or time off duty. Source: original calculation using U.S. Bureau of Labor Statistics public ATUS 2023–2024 microdata.