Are social marketing campaigns effective in preventing child abuse and neglect?

NCPC Issues No. 32 – October 2010

Appendix: Methodology and procedure

Search strategy

For this Issues paper, literature searching was performed between August 2009 and December 2009. Additional searching was undertaken in January 2010 to trace campaigns referred to within publications. Literature was collected from the following databases: EbscoHost (SocIndex, Psycinfo, Psycarticles, Psychology and behavioural sciences, Child development and adolescent studies and Family studies); Informit (Family and Society Plus); ERIC; National Criminal Justice Reference Service; Social Care Online; Proquest; Google and Google Scholar. The Australian Institute of Family Studies library was also systematically searched.

Key terms were entered in a sequential order into each database. Due to the large number of publications identified when using the terms "social marketing" and "media campaign" alone, the search was narrowed with "child abuse or neglect or maltreatment". This resulted in such a small number of hits, the terms "community education" and "community attitudes" were introduced to widen the evidence-base on child maltreatment and parenting. This experience reflected that of Stead et al. (2007), who found the search label "social marketing" to be insufficient for finding literature in databases on public health campaigns specifically targeting physical activity.

Note that the key terms did not specifically include domestic violence, drug use or mental health. Consequently, campaigns associated with these social issues that emerged from the literature search were found in relation to child maltreatment and were selected because the content of the campaign included a message about the effects of these problems upon children or parenting more generally.

The key terms for this systematic literature search were:

  • social marketing + evaluation
  • (social marketing + evaluation) + (child abuse or neglect or maltreatment)
  • (social marketing + evaluation) + parent*
  • (social marketing + evaluation) + child protection
  • media campaign + evaluation
  • (media campaign + evaluation) + (child abuse or neglect or maltreatment)
  • (media campaign + evaluation) + parent*
  • (media campaign + evaluation) + child protection
  • community education + (child abuse or neglect or maltreatment)
  • community education + parent*
  • community education + child protection
  • community attitudes + (child abuse or neglect or maltreatment)
  • community attitudes + parent*
  • community attitudes + child protection

Inclusion/exclusion criteria

The database searches produced 173 publications (journal articles, books, government reports, etc.) about social marketing campaigns. Ten publications were discarded because they did not comprise research on specific social marketing campaigns (e.g., general social marketing literature and opinion pieces). Seventeen publications were excluded as they described population-level surveys of attitudes about child abuse, neglect, maltreatment or family issues, which were not undertaken as part of a social marketing or mass media campaign. In order to focus on contemporary social marketing campaigns implemented during the last 15 years (i.e., 1995-2009), a further seventeen publications released prior to 1995 were also excluded. This resulted in a total sample of 129 publications.

The remaining 129 publications were coded into one of two topic areas: child maltreatment campaigns (defined as social marketing campaigns targeting child abuse and neglect and the determinants of abuse and neglect (e.g., parenting problems, parental substance misuse, domestic violence in families and the impact of these on children), and public health (non-parenting or child maltreatment) campaigns. The coding of publications was independently replicated to validate the procedure. Within these topic areas, the publications were further categorised into Australian or international campaigns. Following a preliminary review, the 93 publications relating to public health campaigns (e.g., domestic violence campaigns not related to families or exposure to children; exercise; quitting smoking; road safety) were also excluded. This resulted in a remaining sample of 36 publications, with 21 distinct child maltreatment-related social marketing campaigns that formed the evidence base for this project.