Table 7.

Unstandardized coefficients from multilevel effects of daily steps on lagged (next-day) energy (N=75)^a.

Outcome	Model 1b: next-day energy			Model 2c: next-day energy				Model 3d: next-day energy
	β	SE	P value	β	SE	P value	β		SE	P value
Intercept	2.01	1.40	.16	1.99	1.40	.16	2.03		1.40	.15
Prior-day steps	.01	0.02	.75	.03	0.07	.71	–.11		0.15	.48
Prior-day energy	–.24	0.03	<.001	.24	0.03	<.001	.24		0.03	<.001
Age (years)	–.001	0.02	.93	–.002	0.02	.93	–.002		0.02	.92
Sex	–.47	0.30	.12	–.46	0.30	.12	–.47		0.30	.12
Condition	–.16	0.27	.55	–.17	0.27	.54	–.16		0.27	.55
Education	.07	0.05	.15	.07	0.05	.15	.07		0.05	.15
Health	.03	0.01	<.001	.03	0.01	<.001	.03		0.01	<.001
Average steps	.01	0.06	.81	.01	0.06	.80	.02		0.06	.80
Prior-day steps×sex	—e	—	—	–.01	0.04	.76	—e		—	—
Prior-day steps×age	—e	—	—	—e	—	—	.002		0.002	.45

^aDaily steps were rescaled by dividing the number of steps by 1000; therefore, the β value reflects change in energy per 1000-step increase.

^bModel 1: level 1 variance=1.11 (SD 1.06); level 2 variance=2.03 (SD 1.43). Akaike information criterion=4394.8; Bayesian information criterion=4450.7; log likelihood=–2186.4.

^cModel 2: level 1 variance=1.11 (SD 1.05); level 2 variance=2.03 (SD 1.43). Akaike information criterion=4396.8; Bayesian information criterion=4457.7; log likelihood=–2186.4.

^dModel 3: level 1 variance=1.12 (SD 1.06); level 2 variance=2.03 (SD 1.43). Akaike information criterion=4396.3; Bayesian information criterion=4457.2; log likelihood=–2186.1.

^eOutcome was not included in the model.