10.7: Measuring Social Influences

Questionnaires are the most common tool for measuring social influences such as social support, parental influences, and group cohesion. A common social support questionnaire is called the Social Support for Exercise Survey that assesses support received from family and friends (Sallis et al., 1987). Example questions include whether family or friends: exercised with me, discussed exercise with me, and helped plan activities around my exercise. A newer measurement tool called the Physical Activity and Social Support Scale has been developed to assess different types of social support (Golaszewski & Bartholomew, 2019). This measure includes separate subscales that capture emotional, companionship, instrumental, informational, and validation support. For parental influence, a new measurement tool called the Parenting Practice Item Bank has been created to capture the three conceptualized parenting strategies described earlier, namely autonomy support, structure, and neglect and control (Mâsse et al., 2020). One of several tools to assess cohesion is called the Group Environment Questionnaire, which assesses the four dimensions of cohesion in a variety of sport and exercise groups (Carron et al., 2002). Although questionnaires are frequently used, there are some limitations of this form of assessment, including a potential self-report bias.

There have been several emerging strategies for assessing social influences including social network approaches (e.g., Prochnow et al., 2020), direct observation of social influences (e.g., Collins & Feeney, 2004; Herbison et al., 2020), and using “big data” from databases (e.g., Carpenter & Amaravadi, 2019). Social network approaches involve individuals describing the networks they are involved in (Prochnow et al., 2020). This may involve having individuals report about people in their networks (e.g., report on family, friends, classmates) or surveying an entire social network (e.g., a classroom at school,or a group exercise class). These networks can be evaluated for a variety of characteristics such as the number of connections a person has, the number of connections within a network, and similarity of the people in a grouping (Prochnow et al., 2020). In a review of 28 social network studies, physical activity seemed to be an attribute that was shared within a grouping such that if a child’s social network was physically active,they also were more likely to be active (Prochnow et al., 2020). A social network approach can be used to look at how interactions across a network of peers or family members influence physical activity behavior.

Another way to measure social influences is to use direct observations where interactions are recorded and then coded to reflect the type of influence. To gain a more objective measure of social support, researchers observed couples interacting and coded the behavior in terms of support provided (Collins & Feeney, 2004). In this study, researchers coded the type of support such as emotional support, and instrumental support in speech writing task. Audio recordings can also be used to capture the social interactions between individuals. Using an electronically activated recorder (EAR), researchers were able to explore youth hockey players’ social interactions in a natural environment (Herbison et al., 2020).Recordings of interactions were then coded to classify the recordings into different types of interactions (Herbison et al., 2020). The use of observation and coding provides a more objective measure for classifying the interactions and, unlike a questionnaire, does not rely on what individuals recall or report.

Another way to gather information on social influences is to use “big data” from databases that exist through social network applications and wearable devices. One study used registered users from a web-based application that was used for creating fitness challenges, allowing people to connect with other users, participating in employer incentivized programs, or sharing results with others (Carpenter & Amaravadi, 2019). Both the number of friends reporting exercise as well as the average amount of friends’ exercise were related to an individual’s own reported level of exercise (Carpenter & Amaravadi, 2019). Others also used data from the Fitbit™ databases and application to look at the social networks on those applications (Stück et al.,2017). Being more active was associated with having a greater number of social ties on the network as well as those ties having a higher average activity level (Stück et al., 2017).These device-based measures and “big” datasets present an interesting avenue for future research examining social influences for physical activity, especially as the use of wearables such as smart watches increases.

The most common tools for measuring social influences are self-reported questionnaires such as scales for social support. Recently, there have been innovations in how to measure social influence including social network analyses, direct observation, as well as using big data. This increase in the diversity of ways to assess social influences will expand our understanding of how these relationships influence our physical activity behavior and our health.