tempted you are not to exercise in the following situations.. .’. The answers include
‘when I am angry’ and ‘when I am satisfied’ to reflect ‘affect’ and ‘when I feel lazy’
and ‘when I am busy’ to reflect competing interests. The authors argue that
such temptations are central to understanding exercise uptake and should be used
alongside the stages of change model.Prospective research
This has examined which factors predict the uptake of exercise. It has often been carried
out in the context of the development of exercise programmes and studies of adherence
to these programmes. Sallis et al. (1986) examined which factors predicted initiation
and maintenance of vigorous/moderate exercise for one year. The results indicated
that exercise self-efficacy, attitudes to exercise and health knowledge were the best pre-
dictors. In a further study, Jonas et al. (1993) followed up 100 men and women and
reported the best predictors of intentions to participate in the exercise programmes
and actual participation were attitudes to continued participation, perceived social
norms and perceived behavioural control. Jones et al. (1998) also examined the pre-
dictors of uptake and adherence, and used repertory grids to explore the personal con-
structs of those individuals who had been referred to exercise as part of an exercise
prescription scheme. They concluded that having realistic aims and an understanding
of the possible outcomes of a brief exercise programme were predictive of adherence to
the programme.
To further understand the predictors of exercise adherence, social cognition models
have been used. Riddle (1980) examined predictors of exercise using the theory of
reasoned action (Fishbein and Ajzen 1975; see Chapter 2) and reported that attitudes
to exercise and the normative components of the model predicted intentions to
exercise and that these intentions were related to self-reports of behaviour. The theory
of planned behaviour (TPB) has also been developed to assess exercise behaviour. Valois
et al. (1988) incorporated a measure of past exercising behaviour (a central variable
in the TPB) and reported that attitudes, intentions and past behaviour were the best
predictors of exercise. Similarly, Hagger et al. (2001) used the TPB to predict exercise in
children and concluded that most variables of the TPB were good predictors of
behavioural intentions and actual behaviour at follow-up. The use of TPB to predict
exercise is discussed further in Focus on research 7.2, opposite. Research has also used
the health belief model (Sonstroem 1988) and models emphasizing exercise self-efficacy
(e.g. Schwarzer 1992), task self efficacy and scheduling self efficacy (Rodgers et al.
2002).
Research has also applied the stages of change model to exercise behaviour (see
Chapters 2 and 5). This model describes behaviour change in five stages: precontem-
plation, contemplation, preparation, action and maintenance (e.g. DiClemente and
Prochaska 1982) and suggests that transitions between changes is facilitated by a cost
benefit analysis and by different cognitions. Marcus et al. (1992) examined the relation-
ship between the pros and cons of exercise and stage of change in 778 men and women.
The pros and cons of exercise and decisional balance (pros versus cons) was related to
exercise adoption and that higher ratings of pros were found in those individuals closer178 HEALTH PSYCHOLOGY