Association between time preference, present-bias and physical activity: implications for designing behavior change interventions

Table 3 The association between time preferences and physical activity (100 mins/week) and socio-demographics

Variables	Coefficient	95% CI	Standard error	t-value
Dependent variable: Discount rate (β)
PAL minutes	− 0.091^b	[−0.171,-0.012]	0.041	−2.26
Old	−0.036	[−0.116,0.044]	0.041	−0.88
Male	0.057	[−0.016,0.130]	0.037	1.54
Household income	− 0.039	[− 0.091,0.013]	0.026	−1.47
Single	−0.138^b	[−0.263,-0.013]	0.064	−2.17
Child	−0.134^c	[− 0.217,-0.051]	0.042	−3.17
Own house	0.097^b	[0.007,0.186]	0.046	2.12
Constant	0.339^c	[0.114,0.563]	0.114	2.96
Dependent variable: Present-biasedness (α)
PAL minutes	−3.790^c	[−5.990,-1.591]	1.122	−3.38
Old	−2.607^c	[−3.948,-1.266]	0.684	−3.81
Male	1.296	[−0.277,2.869]	0.803	1.61
Household income	−0.888^a	[−1.796,0.020]	0.463	− 1.92
Single	−3.619^b	[−6.697,-0.542]	1.570	−2.31
Child	−2.533^c	[−4.058,-1.008]	0.778	−3.26
Own house	5.537^c	[2.398,8.676]	1.601	3.46
Constant	5.745^b	[0.959,10.531]	2.442	2.35
Dependent variable: Risk preference (γ)
PAL minutes	−0.102^c	[− 0.177,-0.027]	0.038	−2.68
Old	−0.007	[−0.096,0.082]	0.046	−0.16
Male	0.003	[−0.069,0.075]	0.037	0.08
Household income	0.003	[−0.038,0.044]	0.021	0.16
Single	0.065	[−0.029,0.160]	0.048	1.35
Child	−0.014	[−0.103,0.076]	0.046	−0.30
Own house	0.060	[−0.069,0.189]	0.066	0.91
Constant	0.770^c	[0.452,1.087]	0.162	4.74
μ_RA	0.145^c	[0.085,0.205]	0.031	4.76
μ_DR	0.099	[−0.076,0.274]	0.089	1.11
Log likelihood	− 3371.89
Number of observations	6650

Notes: PAL minutes is measured as per 100 min of physical activity per week across the 12-week intervention period; Old equals 1 if the participants is older than 40, otherwise 0; Male takes the value of 1 if the participant is male, 0 otherwise; Household income takes five values: 1 (below £15,000), 2 (£15,000 - £29,999), 3 (£30,000 - £ 49,999), 4 (£50,000 - £ 79,999), 5 (£ 80,000 or more); Single equals 1 if the participant is single, otherwise 0; Child takes the value of 1 if the participant has at least one child, otherwise 0; Own house takes the value of 1 if the participant owns a house, otherwise 0; ^aindicates 10%, ^bindicates 5% and ^cindicates 1%. Standard errors are clustered at the individual level

ISSN: 1471-2458