Objective To investigate the effects of two Internet-based weight loss interventions on physical activity (PA) and dietary behaviors using two approaches for computing combined behavior change. Method Participants were 352 overweight/obese women and men completing 12-month interventions in San Diego, California during 2002–2007. Moderate-to-vigorous PA (MVPA) and sedentary time were measured with accelerometers, and dietary fat and fruit and vegetable intake were assessed with food frequency questionnaires. Longitudinal analyses tested the effect of the intervention on combined health behavior change quantified using a standardized residualized change index (SRCI) and a risk factor change index (RFCI). Results At baseline, participants engaged in an average of 153 min/week of MVPA and 525 min/day of sedentary time, and consumed 37% of calories from fat and qqqlt; 3 fruits and vegetables per day. The interventions had a significant effect on combined behavior change as measured with each approach (p qqqlt; 0.001). The intervention effect was larger (p qqqlt; 0.001) when evaluated using the SRCI (standardized regression coefficient [Beta] = 0.30) than the RFCI (β = − 0.18). Conclusion Interventions that target both PA and dietary behaviors appear effective. The SRCI was more sensitive for evaluating the intervention, but the RFCI may be easier to use for communicating public health significance.