Table 1
Demographic information of subjects in Grades 10, 11, and 12.
| Grade | Sample Size | Covariates | n (%) or Mean (SD) | |
|---|---|---|---|---|
| 10 | N = 95 | Race * | 1: Hispanic | 35 (36.84%) |
| 2: African American | 25 (26.32%) | |||
| 3: Asian | 2 (2.11%) | |||
| 4: White | 33 (34.74%) | |||
| BMI Group * | 0: Low BMI Group | 76 (80.00%) | ||
| 1: High BMI Group | 19 (20.00%) | |||
| BMI ** | 25.50 (5.44) | |||
| Family Affluence * | 1: Low Affluence | 30 (31.58%) | ||
| 2: Moderate Affluence | 43 (45.26%) | |||
| 3: High Affluence | 22 (23.16%) | |||
| Parental Education ** | Parental Education 1 | 3.49 (1.92); n = 91 | ||
| Parental Education 2 | 3.13 (1.84); n = 60 | |||
| 11 | N = 85 | Race * | 1: Hispanic | 30 (35.29%) |
| 2: African American | 25 (29.41%) | |||
| 3: Asian | 1 (1.18%) | |||
| 4: White | 29 (34.12%) | |||
| BMI Group * | 0: Low BMI Group | 68 (80.00%) | ||
| 1: High BMI Group | 17 (20.00%) | |||
| BMI ** | 25.39 (5.32) | |||
| Family Affluence * | 1: Low Affluence | 26 (30.59%) | ||
| 2: Moderate Affluence | 38 (44.71%) | |||
| 3: High Affluence | 21 (24.71%) | |||
| Parental Education ** | Parental Education 1 | 3.53 (1.98); n = 81 | ||
| Parental Education 2 | 3.21 (1.84); n = 53 | |||
| 12 | N = 84 | Race * | 1: Hispanic | 29 (34.52%) |
| 2: African American | 23 (27.38%) | |||
| 3: Asian | 2 (2.38%) | |||
| 4: White | 30 (35.71%) | |||
| BMI Group * | 0: Low BMI Group | 68 (80.95%) | ||
| 1: High BMI Group | 16 (19.05%) | |||
| BMI ** | 25.15 (5.38) | |||
| Family Affluence * | 1: Low Affluence | 26 (30.95%) | ||
| 2: Moderate Affluence | 39 (46.43%) | |||
| 3: High Affluence | 19 (22.62%) | |||
| Parental Education ** | Parental Education 1 | 3.53 (1.92); n = 80 | ||
| Parental Education 2 | 3.09 (1.84); n = 53 | |||
[i] * n (%) is the number (percentage) for each level of categorical covariate.
** Mean (SD) is the mean (standard deviation) of each continuous covariate. Parental Education 1 and 2 describe the highest level of education completed by each respective guardian in the household; education level is reported using a seven point scale.

Figure 1
Activity patterns and activity differences of subject 909010214. In the left-hand side plots (a), (b), and (c), activity counts are shown by black dots. In the right-hand side plots (d), (e), and (f), the differences of activity counts between the three pairs of grades are shown by black dots. The smoothed Fourier expansion in each plot is shown by a red solid curve. Abbreviation: Diff = Difference.

Figure 2
Smoothed activity patterns. In the plots (a), (b), and (c), the individual smoothed Fourier expansions of grades 10, 11, and 12 are shown respectively. In plot (d), smoothed Fourier expansions of all three grades are shown.

Figure 3
Observed F-statistic and t-statistic results and related permutation critical values for a relationship between activity counts and grades. The red solid curve represents the observed statistic F(t) or T(t) at each time point, and the green dashed and blue dotted lines correspond to permutation critical values for the maximum and point-wise statistics at significance level α = 0.05, respectively.

Figure 4
Smoothed activity patterns by BMI status in the three grades. In the plots (a) and (b), the individual smoothed Fourier expansions of grade 10 are shown for low-BMI vs. high-BMI groups, respectively. In the plots (c) and (d), the individual smoothed Fourier expansions of grade 11 are shown for low-BMI vs. high-BMI groups, respectively. In the plots (e) and (f), the individual smoothed Fourier expansions of grade 12 are shown for low-BMI vs. high-BMI groups, respectively.

Figure 5
Observed t-statistic results and related permutation critical values for the activity counts between low-BMI and high-BMI groups in the three grades. The red solid curve represents the observed statistic T(t) at each time point, and the green dashed and blue dotted lines correspond to permutation critical values for the maximum and point-wise t-statistics at significance level α = 0.05, respectively.
Table S.1
Sub-sample Sizes in Grades 10, 11, and 12 by Recording Months.
| Recording Month | Sub-sample Sizes | ||
|---|---|---|---|
| Grade 10 | Grade 11 | Grade 12 | |
| 2 | 0 | 13 | 22 |
| 3 | 0 | 18 | 31 |
| 4 | 0 | 33 | 16 |
| 5 | 0 | 10 | 15 |
| 6 | 31 | 9 | 0 |
| 7 | 54 | 2 | 0 |
| 8 | 10 | 0 | 0 |
| Total | 95 | 85 | 84 |
Table S.2
Sub-sample Sizes in Grades 10, 11, and 12 by Summer Recording (i.e., Months 6, 7, 8) vs. Non-summer Recording (i.e., Months 1–5, 9–12).
| Summer vs. Non-Summer | Sub-sample Sizes | ||
|---|---|---|---|
| Grade 10 | Grade 11 | Grade 12 | |
| Summer | 95 | 11 | 0 |
| Non-summer | 0 | 74 | 84 |
| Total | 95 | 85 | 84 |
Table S.3
Sub-sample Sizes in Grades 10, 11, and 12 by Recording Months, and Low BMI and High BMI Groups.
| The Recording Month | Sub-sample Sizes | |||||
|---|---|---|---|---|---|---|
| Grade 10 | Grade 11 | Grade 12 | ||||
| Low BMI | High BMI | Low BMI | High BMI | Low BMI | High BMI | |
| 2 | 0 | 0 | 2 | 11 | 5 | 17 |
| 3 | 0 | 0 | 3 | 15 | 8 | 23 |
| 4 | 0 | 0 | 7 | 26 | 3 | 13 |
| 5 | 0 | 0 | 3 | 7 | 0 | 15 |
| 6 | 4 | 27 | 2 | 7 | 0 | 0 |
| 7 | 13 | 41 | 0 | 2 | 0 | 0 |
| 8 | 2 | 8 | 0 | 0 | 0 | 0 |
| Total | 19 | 76 | 17 | 68 | 16 | 68 |
Table S.4
Sub-sample Sizes in Grades 10, 11, and 12 by Summer Recording (i.e., Months 6, 7, 8) vs. Non-summer Recording (i.e., Months 1–5, 9–12), and Low BMI and High BMI Groups.
| The Recording Month | Sub-sample Sizes | |||||
|---|---|---|---|---|---|---|
| Grade 10 | Grade 11 | Grade 12 | ||||
| Low BMI | High BMI | Low BMI | High BMI | Low BMI | High BMI | |
| Summer | 19 | 76 | 2 | 9 | 0 | 0 |
| Non-summer | 0 | 0 | 15 | 59 | 16 | 68 |
| Total | 19 | 76 | 17 | 68 | 16 | 68 |

Figure S.1
Observed F-statistic results and related permutation critical values for a relationship between activity counts and race/ethnicity. The red solid curve represents the observed statistic F(t) at each time point, and the green dashed and blue dotted lines correspond to permutation critical values for the maximum and point-wise statistics at significance level α = 0.05, respectively.

Figure S.2
Observed F-statistic results and related permutation critical values for a relationship between activity counts and family affluence. The red solid curve represents the observed statistic F(t) at each time point, and the green dashed and blue dotted lines correspond to permutation critical values for the maximum and point-wise statistics at significance level α = 0.05, respectively.
