Ridgeline plot with a barcode strip. I chose a ridgeline plot because the question is about how a whole distribution shifts across eleven ordered age groups, and stacking one density curve per group lets the eye follow the peak as it slides left with age, which a single average per group would hide. The x-axis is total movement per day in MIMS (Monitor-Independent Movement Summary) units from the wrist accelerometer, from 0 to 35k, shared by every ridge so positions can be compared directly. The y-axis has no numeric scale: each row is one age group, labeled on the left from 3–5 at the top to 70–80 at the bottom, and ridge height shows only where people are concentrated within that group. Under each ridge, a barcode strip draws one faint tick per person, so readers see the 6,785 actual people rather than only a smoothed curve, and a black tick marks each group's median. A small legend at the top right explains the three marks, and n is printed beside every row. I used a single green for all ridges, the same green that means movement in Visualization 2, because age is already encoded by vertical position and a second color would add nothing; green signals that more movement is the healthy direction. The chart answers the question clearly: movement is highest at ages 6–8 (median about 20,300 MIMS), drops about 31% by ages 15–19, stays fairly level from the 20s to the 50s, and falls again to about 10,000 by 70–80, when only 10% of people move more than the typical 20–29-year-old. Uneven band widths (3 years for children, 10 for adults) mean the y-axis is not a linear age scale, and smoothing can blur small bumps in the distribution.
Person-by-hour heatmap, standardized within each person. A heatmap fits this question because it involves two dimensions at once, clock time and age, and it can show every one of the 6,785 people without averaging them into groups: each row is one person and each of the 24 columns is one clock hour. Raw movement counts are not comparable for this question, because a young child moves about twice as much in a day as an older adult, so raw values would mostly show who moves more rather than when. I therefore standardized each row to the person's own day: each cell is the share of that person's 24-hour movement that fell in that hour, so every row sums to 100% and a perfectly even day would be about 4% per hour. The x-axis runs from midnight to midnight with labels every three hours. The y-axis stacks people from youngest (age 3, top) to oldest (80+, bottom), with tick labels and thin white rules where each age band begins; within each age, rows are sorted by the person's average time of movement, so early movers sit above late movers. Color is the only magnitude channel, so I used a single-hue green scale from near-white (0%) to deep green (10% or more of the day in one hour), with a labeled color bar at the top; green marks the positive end of the measure, and one hue reads naturally as "more" without implying categories. The answer is in how the dark band shifts: children's movement is spread from morning into the early evening and is mostly over by 9 pm (a median of 6.6% of their movement falls after 9 pm); from about age 12 the band extends toward midnight (11.2% after 9 pm for ages 12–19); and for people 60 and older the band shifts earlier, with a median of 33.9% of the day's movement between 6 am and noon. The heatmap shows timing, not amount, so it hides how much less older adults move overall, which Visualization 1 shows. Averaging each hour across a person's valid days also blends weekdays and weekends.
Scatter plot of every person, with a mean line for each sex. I chose a scatter plot because both variables are quantitative and I wanted to show the shape of growth and decline without hiding how much people vary at each age. Each of the 7,807 dots is one person's strongest grip-dynamometer trial. The x-axis is age in years from 6 to 80+, with dots spread randomly within each year (jitter) so people of the same age do not stack into one vertical line. The y-axis is grip strength in kilograms, starting at zero so differences are not exaggerated, with light gridlines every 10 kg. Color encodes sex, orange for female and blue for male, and the legend gives each group's size. The dots are small and semi-transparent so dense regions appear darker. On top, a thicker line in the same color traces each sex's mean grip at every year of age, with a thin white outline so it stays readable over the dots; the lines make the point of divergence visible without any added text. The chart shows the two lines running together until age 12 (mean 23.9 kg for girls and 24.1 kg for boys) and separating sharply from age 13, when boys' mean grip jumps ahead; by 18 the gap is about 16 kg. Men's mean grip peaks in the early-to-mid 30s at about 52 kg, women's at about 34 kg, and both decline after the 50s. The dots also show a very wide spread at every age: about 12% of men aged 70 and older grip harder than the weakest quarter of men in their 30s. Single-year means are noisy, which is why the lines zigzag, and grip measures only one muscle group.
Together, the three visualizations show that the lifespan changes movement in three different ways. The first shows the amount: daily movement peaks in early childhood, drops sharply through adolescence, and declines again after 60. The second shows the timing: as children become teenagers, a larger share of their movement shifts later into the night, and in later life it concentrates earlier in the day. The third shows the physical capacity behind movement, and it follows a different curve: grip strength rises through adolescence, with boys pulling away from girls at about age 13, peaks in the early-to-mid 30s, long after daily movement has already peaked, and declines later and more gradually. No single chart reveals that the most active stage of life and the strongest stage of life are more than two decades apart. Together they suggest that programs to keep people moving should start in adolescence, when movement drops most sharply, while strength-focused programs matter most after 60. What remains unanswered is cause: these are measurements of different people at different ages at one point in time, so following the same people over many years would be needed to show how one person's movement and strength change together.