Behavioral Research

Table of Contents
1 Description & Theoretical Background

Using Normative Beliefs in Behavior Change Paradigms


Measurement and Methodological Issues


Factors that Increase the Importance of Normative Beliefs


Related Concepts

6 References
7 Measures Appendix: A
8 Measures Appendix: B
9 Published Examples

Download Full Text (PDF)

Other Constructs



Dispositional Optimism




Illness Representations

  Implementation Intentions
  Intention, Expectation, and Willingness
  Normative Beliefs
  Optimistic Bias
  Perceived Benefits
  Perceived Control
  Perceived Severity
  Perceived Vulnerability
  Self-Reported Behavior
  Social Influence
  Social Support

Normative Beliefs
David Trafimow

<< Previous

2 Using Normative Beliefs in Behavior Change Paradigms

Using normative beliefs in the context of the theory of reasoned action
Historically, there has been a strong tendency for health researchers to use normative beliefs in the context of the theory of reasoned action to predict and influence health behaviors. One strategy has been as follows. Because the measurement of behavioral intentions, attitudes, and subjective norms does not require any preliminary research and can be done with a simple questionnaire (see Ajzen & Fishbein, 1980 and Trafimow, 2004 for details), a resource-saving first step is to measure these variables in a sample that represents the population of interest. After performing a multiple regression analysis, and obtaining beta weights indicating the relative contributions of attitudes and subjective norms to predict behavioral intentions, one can easily determine which of the two variables contributes most to this prediction. If attitudes are a good predictor and subjective norms are not, then there is no point in wasting resources on the normative pathway (which means there is no point in performing elicitation studies to find out about relevant normative beliefs because they do not matter anyway). Instead, the focus should be on the attitudinal pathway. On the other hand, if subjective norms are a good predictor of behavioral intentions, then this is a strong indication that it is worthwhile to take the normative pathway seriously. In this case, the next step would be to perform an elicitation study to find out the relevant normative beliefs. This can be done by simply asking participants to list others who are important to them in deciding whether or not to perform the behavior of interest. Once the relevant normative beliefs have been elicited, they can be measured in the main study (see below for a description).

A second strategy is to perform elicitation studies right away, to find out the relevant normative and behavioral beliefs. Then, using the data from the elicitation study as a basis, a questionnaire is constructed that includes all of the theory of reasoned action variables for use in the main study. The advantage of the second strategy is that it requires fewer steps (two steps instead of three). A disadvantage is that, because there is no a priori way to know whether both the attitudinal and normative pathways are relevant to the behavior of interest in the particular population of concern, both the elicitation and main studies have to include both pathways, and are consequently more complex.

Regardless of which strategy is used, at least two types of useful information can be gleaned. In the first place, multiple regression analysis can be used to determine whether the attitudinal or normative pathway is most important for the behavior or population of interest. Secondly, for the purposes of intervention, those normative beliefs (or motivations to comply) that are most predictive of subjective norms (or behavioral intentions or behaviors) can be determined so that they can be the focus of an intervention.

Using normative beliefs in the context of the theory of planned behavior
Ajzen's (1988) theory of planned behavior is similar to Fishbein's theory of reasoned action, but with the addition of perceived behavioral control-the extent to which a behavior is believed to be under the person's control. Therefore, instead of there being two causal pathways to behavior as in the theory of reasoned action, there are three. These are the attitudinal, normative, and control pathways. However, the way normative beliefs are used in the theories of reasoned action and planned behavior are similar.

Using normative beliefs in the context of within-participants research paradigms
Trafimow (1994) pointed out that there might be different causal pathways to behaviors for different persons. For example, some people might be more prone to perform behaviors on the basis of the attitudinal pathway whereas the normative pathway might be more important for other people. Consequently, intervention strategies should be focused not only on the behavior of interest, but also on the specific persons of interest. To do this, however, it is necessary to have a method of determining, for every person, whether his or her behaviors tend to be caused by attitudinal or normative factors. Trafimow and his colleagues (Trafimow & Finlay, 1996; Trafimow, Kiekel, & Clason, 2004) showed that there is such a method. The idea is to consider a large set of behaviors in a particular domain of interest rather than a single behavior. For example, the domain could be cancer prevention behaviors, exercise behaviors, safety behaviors, dietary behaviors, and so on. Instead of measuring variables (e.g., theory of reasoned action variables) for only one behavior, the researcher measures these variables for the whole set of behaviors. This research design makes it possible to conduct traditional between-persons analyses (these are analyses within a single behavior and across persons), but it is also possible to perform within-persons analyses (these are analyses within a single person and across behaviors). Thus, it is possible to determine, for each person, whether his or her behaviors are more controlled by the attitudinal or normative pathway. Finlay and her colleagues (Finlay, Trafimow, & Moroi, 1999; Finlay, Trafimow, & Villarreal, 2002) have demonstrated the usefulness of this approach in a variety of health domains, and Sheeran, Trafimow, Finlay, and Norman (2002)  have demonstrated that the approach can be extended to include additional variables such as perceived behavioral control.

<< Previous

Search | Help | Contact Us | Accessibility | Privacy Policy

DCCPSNational Cancer Institute Department of Health and Human Services National Institutes of Health

Health Behavior Constructs: Theory, Measurement, & Research  You can Quit smoking Now!