Social Influence
Thomas A. Wills, Michael G. Ainette, and Carmella Walker

There is little question that social influence is an important determinant of health-related behaviors, including not only cigarette smoking and alcohol use but a range of other behaviors as well. Thus the basic question for new research on health behavior is not necessarily whether social influence should be studied, but what aspect should be measured. This section provides a few simple suggestions about research in the area, to augment the investigator's reading of primary literature and theory in his/her area of interest.

Including multiple measures. It can be suggested that multiple measures, assessing different aspects of social influence, are preferable to a single measure. If assessment space is limited and actuarial prediction is the only research goal, then items indexing peer behavior (smoking, alcohol, etc.) are a reasonable choice because they typically have strong correlations with respondent behavior. However a single measure of peer behavior may provide little understanding of how social influences operate: for example is it through explicit pressure, normative perceptions, or social image processes? For this reason it is desirable to include two or more measures of social influence processes, in order to understand more about how social networks influence individual behaviors.

Choosing the dimensions. A number of social measures are potentially available so a researcher needs to consider which ones may be most relevant for a given population. For example measures of peer influences may be very relevant for younger persons who commonly spend their time in large school populations, but less relevant for older samples where primary networks are more important. Social processes may operate differently for behaviors that are common vs. rare in the general population, and researchers should consider how social influence mechanisms may differ in these conditions. Similarly, asking about whether the health-related behavior is simple or complex to adopt suggests ancillary questions about what aspects of the social environment may be involved in either supporting or deterring the adoption (or cessation) of the behavior. Through asking these kinds of questions, the investigator may construct a preliminary model of the causal chains that produce or inhibit the behavior, and this kind of thinking can suggest what measures will be most relevant for the research.

Doing descriptive research. Existing measures of general social influence processes may be supplemented with descriptive studies to gather detailed information about social processes as they occur in the investigator's population. For example qualitative studies of adolescent smoking have been informative in providing perspectives about smoking onset that differ somewhat from common expectations about why teens smoke (see Friedman, Lichtenstein, & Biglan, 1985 xClose
Friedman, L. S., Lichtenstein, E., & Biglan, A. (1985). Smoking onset among teens: An empirical analysis of initial situations. Addictive Behaviors, 10, 1-13.; Lucas & Lloyd, 1999 xClose
Lucas, K., & Lloyd, B. (1999). Starting smoking: Girls' explanations of the influence of peers. Journal of Adolescence, 22, 647-655.; Sussman et al., 1993 xClose
Sussman, S., Hahn, G., Dent, C. W., & Stacy, A. W. (1993). Naturalistic observation of adolescent tobacco use. International Journal of the Addictions, 28, 803-811.). Preliminary research using individual open-ended interviews for focus groups may be helpful for evaluating the applicability of proposed measures and suggesting new facets or dimensions that are not precisely represented in existing scales.

Thinking about subgroups. The measures described in this section were validated in epidemiologic research but there is always the possibility that social influence processes may have differential impact in subgroups of the population. It is desirable for investigators to consider what they know about a given population (students, community residents, or clinic patients) and anticipate how measures might be focused or adapted for these persons. For example there is some evidence that peer influences processes may differ by ethnicity (Landrine, Richardson, Klonoff, & Flay, 1994 xClose
Landrine, H., Richardson, J., Klonoff, E., & Flay, B. (1994). Cultural diversity in the predictors of adolescent cigarette smoking: The relative influence of peers. Journal of Behavioral Medicine, 17, 331-346.; Vaccaro & Wills, 1998 xClose
Vaccaro, D., & Wills, T. A. (1998). Stress coping factors in adolescent substance use: Ethnic and gender differences in samples of urban adolescents. Journal of Drug Education, 28, 257-282.). Hence in planning a new study the investigator may use existing knowledge about the population to consider what types of subgroup effects (by gender, age, ethnicity, etc.) there may be in the population and to consult literature on cultural variables (e.g., Catalano, Hawkins, Krenz, Gillmore, Morrison, Wells, & Abbott, 1993 xClose
Catalano, R.F., Hawkins, J.D., Krenz, C., Gillmore, M., Morrison, D., Wells, & Abbott, R. (1993). Using research to guide culturally appropriate drug abuse prevention. Journal of Consulting and Clinical Psychology, 61, 804-811.). As suggested above, qualitative research and pilot studies may also be useful in this regard.

Considering the larger environment. Recent research has drawn attention to the effect of larger environments on health and illness, (Diez-Roux et al., 2003 xClose
Diez-Roux, A. V., Merkin, S., Hannan, P., Jacobs, D. R., & Kiefe, C. I. (2003). Area characteristics, individual-level socioeconomic factors, and smoking in young adults. American Journal of Epidemiology, 157, 315-326.; Gibbons et al., 2004 xClose
Gibbons, F. X., Gerrard, M., Cleveland, M. J., Wills, T. A., & Brody, G. H. (2004). Perceived discrimination and substance use in African-American parents and their children: A panel study. Journal of Personality and Social Psychology, 86, 517-529.; Kawachi et al., 1997 xClose
Kawachi, I., Kennedy, B. P., Lochner, K., & Prothrow-Stith, D. (1997). Social capital, income inequality, and mortality. American Journal of Public Health, 87, 1491-1498.; Lee & Cubbin, 2002 xClose
Lee, R. E., & Cubbin, C. (2002). Neighborhood context and youth cardiovascular health behaviors. American Journal of Public Health, 92, 428-436.). Investigators planning research on health-related behavior might be advised to consider including measures of environmental-level variables relevant to the behaviors they wish to study, either measures of specific environments for younger persons (Novak & Clayton, 2001 xClose
Novak, S. P., & Clayton, R. R. (2001). The influence of school environment and self-regulation on transitions between stages of cigarette smoking. Health Psychology, 20, 196-207.) or measures of larger social environments for research with adults (Brody et al., 2001 xClose
Brody, G. H., Ge, X., Conger, R., Gibbons, F. X., Murry, V. M., Gerrard, M., & Simons, R. L. (2001). The effect of neighborhood disadvantage, collective socialization, and parenting on African-American children's association with deviant peers. Child Development, 72, 1231-1246.). The measures are relatively brief ones and can provide an increment in predictive power, above and beyond the contribution of individual-level characteristics, so the research balance can be a favorable one.