Most theories designed to understand and predict health behaviors–including protection motivation theory (PMT, Rogers, 1983), the prototype/willingness model (PWM; Gibbons, Gerrard, & Lane, 2003), the theory of planned behavior (TPB; Ajzen, 1991), and social cognitive theory (Bandura, 1997)—construe the formation of a goal intention as the key act of willing that promotes goal attainment. Goal intentions can be defined as the instructions that people give themselves to perform particular behaviors or to achieve certain desired outcomes (Triandis, 1980) and are measured by items of the form, "I intend to achieve X!" Goal intentions can vary in strength as they index a commitment to pursuing a goal or performing a behavior (Gollwitzer, 1990; Webb & Sheeran, 2005). For example, smokers may have weak intentions to quit smoking next week but strongly intend to quit ‘some day;’ a woman may intend to get a mammogram soon, and an overweight man might definitely intend to lose a certain amount of weight during the coming year.
Correlational surveys that measure participants’ goal intentions at one time-point and measure behavioral outcomes at a later time-point seem to support the predictive validity of goal intentions. For instance, a meta-analysis of 10 previous meta-analyses found that goal intentions accounted for 28% of the variance in behavior, on average, across 422 studies (Sheeran, 2002). Although R2 = .28 is a large effect size (cf. Cohen, 1992), a substantial proportion of the variance in behavior is not explained by goal intentions. The magnitude of the gap between intentions and action is illuminated by studies that decomposed this relationship in terms of a 2 (goal intention: to act vs. not to act) X 2 (goal attainment: acted vs. did not act) matrix (Orbell & Sheeran, 1998). A review of health behavior matrices (e.g., condom use, exercise, cancer screening) found that people translated their ‘good’ intentions into action only 53% of the time (Sheeran, 2002). More seriously, evidence indicates that correlational studies overestimate the consistency between intentions and behavior. A meta-analysis of experimental studies that succeeding in changing goal intentions among treatment versus control conditions (Webb & Sheeran, 2006) found that the magnitude of the difference in subsequent behavior was only small-to-medium (R2 = .03). In sum, accumulated evidence indicates that forming even strong goal intentions does not guarantee goal attainment.