A Web-Based Instrument for Evaluating and Planning Health Communication

The self-report measure for evaluating health communications was developed in response to the massive communications efforts surrounding the H1N1 pandemic and the drive to have people get inoculated in order to control the spread of the disease. The concern was particularly evident on college campuses where susceptibility and impact were considered to be high due to large numbers of people coming into close and frequent contact.

Theoretical Background and research questions/hypothesis:  Understanding the drivers to particular actions is the basis for the development of communication messages whether the product or service is a new video game or a new flu vaccine. This study sought to create a valid and reliable scale for gaining information about health beliefs (i.e. H1N1 inoculation). Secondarily, the study aimed to create an instrument that would explore the sources of information surrounding a particular health issue and provide a framework for evaluating a specific health communication campaign. Despite its widespread use as the basis for understanding preventative health beliefs and behaviors, the literature surrounding the use of the Health Belief Model relative to preventative health behaviors includes little on the development of valid and reliable scales. For the few that were found (searching back to 1970) only four included the wording for the items used in the scale. Of these, none related specifically to inoculation. In addition, there was no literature on the use of web-based instruments for such measures.

Methods:  The email invitation was sent to the entire student population of the Charles River Campus of Boston University (N = 27,709) in late March 2010. A reminder email was sent 10 days afterward. The questionnaire was implemented online using a popular online survey service. SPSS was used to conduct data analysis. Construct validity was tested using confirmatory factor analysis. The principal component analysis was conducted using Varimax rotation listwise (n=1021). Cronbach’s alpha was used to test for reliability.

Results:  Twenty-three of the twenty-four items loaded across eight components. All components showed statistically significant levels of reliability (a>.60; p<.05). Data was also obtained on levels of exposure to information about H1N1 through various forms of mass communication.

Conclusions:  Data reveal the drivers for college students to choose to perform single-episode preventative health behavior; in this case, the taking of the shot for H1N1. The data also demonstrates the value of the instrument in its ability to gather specific campaign-related information that can be used for both summative (e.g. levels of recall and awareness, perceptions of messages) and formative purposes (e.g. channels of distribution; messaging; and product design).

Implications for research and/or practice: Organizations rarely have the resources, whether time, money or staffing, to conduct meaningful evaluation. This instrument provides a theory-based, valid, reliable scale that can be easily implemented. That it has been proven as delivered on the web allows its use when there is a potential for a larger number of responses since its per person costs are very low. It also provides a framework for adaptability to evaluate individual health communication campaigns.