Advanced search
722
Views
86
CrossRef citations to date
0
Altmetric
Original Articles

Can we explain why some people do and some people do not act on their intentions?

, , &
Pages 3-18
Published online: 23 Jan 2007

Behavioural theorists have identified attitudes, perceived norms and self-efficacy as the important determinants of people's intentions to engage in a given behaviour. Because intentions predict behaviour, these same variables also account for a considerable amount of the variation in behaviour. Nevertheless, there is often a substantial proportion of the population who do not act on their intentions. While a recently proposed integrative theory of behaviour suggests that these ‘failures’ are due either to a lack of skills and/or to the presence of environmental constraints, it has also been argued that the determinants of intention may have a direct, as well as in indirect, effect on behaviour. This paper uses data from a longitudinal study (Project RESPECT) to explore the extent to which attitudes, perceived norms and self-efficacy explain why some people do and others do not act on their intentions to engage in a health protective behaviour. Although the data provide further evidence that these three variables account for a significant proportion of the variance in intentions (and behaviour), they perform poorly when predicting behaviour for persons with pre-existing high intentions. It may be reasonable to ask whether a ‘new’ theory is needed to explain why some people do, and some people do not, act on their intentions.

Acknowledgements

The Project RESPECT study group: Baltimore: Carolyn Erwin-Johnson, MA; Andrew L. Lentz, MPA; Mary A. Staat, MD, MPH; Dawn Sweet, PhD; Jonathan M. Zenilman, MD (Principal Investigator [PI]). Denver: John M. Douglas, Jr., MD (PI); Tamara Hoxworth, PhD; Ken Miller, MPH; William McGill, PhD. Long Beach: Ruth Bundy, PhD (co-PI); Laura A. Hoyt, MPA; C. Kevin Malotte, DrPH; Fen Rhodes, PhD (PI). Newark: Michael Iatesta, MA; Eileen Napolitano (co-PI); Judy Rogers, MS; Ken Spitalny, MD (PI). San Francisco: Gail A. Bolan, MD (PI); Coleen LeDrew; Kimberly A.J. Coleman; Luna Hananel, MSW; Charlotte K. Kent, MPH. NOVA, Inc.: Robert Francis, PhD (PI); Christopher Gordon; Nancy Rosenshine, MA; Carmita Signes. CDC: Sevgi Aral, PhD; Robert H. Byers, PhD; Beth Dillon, MSW, MPH; Martin Fishbein, PhD; Sandra Graziano, PhD; Mary L. Kamb, MD, MPH; William Killean; James Newhall, PhD; Daniel Newman, MS; Thomas A. Peterman, MD, MSc; Karen L. Willis, RN.

Notes

Although probit regression coefficients in a metric of Z scores (Agresti, 1990, pp. 102 – 104), the probit slopes can be transformed to represent the change in the probability of being in one outcome level versus the other given a one unit change in the predictor value, e.g. the change in probability of being an intender versus a non-intender for a unit change in self-efficacy for an average respondent (Greene, 1993, p. 639). Because of this feature, we emphasize the ‘change in probability’ rather than the ‘change in Z score’ interpretation of the probit results.

 

Related research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.