Eeva M Koutaniemi1, Elina Einiö2,3,4. 1. Tulos Helsinki, Helsinki, Finland. 2. University of Helsinki, Population Research Unit, Department of Social Research, Finland. 3. London School of Economics and Political Science, London, UK. 4. Max Planck Institute for Demographic Research, Rostock, Germany.
Abstract
Aims: Domestic violence is a major health concern and a largely hidden crime. It is estimated that authorities receive information in only a minority of cases. This study investigated seasonal patterns in seeking help for domestic violence by employing Google data. Methods: We utilised monthly Google search data and police calls made in Finland in 2017 to analyse seasonal variation in seeking help for domestic violence. We calculated rate ratios for selected Google terms based on observed search volumes (O) and those expected without seasonal variation (E). These rate ratios (O/E) were compared with the corresponding police call statistics registered as domestic violence. Results: The findings on Google search data showed increased search volumes for domestic violence in November, January and March. The rate ratio (O/E) for searches for shelters is 1.30 in November [95% confidence interval (CI): 1.27-1.33], 1.17 in January (95% CI: 1.14-1.20), and 1.16 in March (95% CI: 1.13-1.29). These peaks in search volumes occur within the same months as those observed in the corresponding police calls categorised as domestic violence. Police data also showed somewhat higher volumes in April. Conclusions: The study suggests that Google search volumes can be used to study the highest peaks in seeking help for domestic violence in countries with a high level of Internet usage and no available police data.
Aims: Domestic violence is a major health concern and a largely hidden crime. It is estimated that authorities receive information in only a minority of cases. This study investigated seasonal patterns in seeking help for domestic violence by employing Google data. Methods: We utilised monthly Google search data and police calls made in Finland in 2017 to analyse seasonal variation in seeking help for domestic violence. We calculated rate ratios for selected Google terms based on observed search volumes (O) and those expected without seasonal variation (E). These rate ratios (O/E) were compared with the corresponding police call statistics registered as domestic violence. Results: The findings on Google search data showed increased search volumes for domestic violence in November, January and March. The rate ratio (O/E) for searches for shelters is 1.30 in November [95% confidence interval (CI): 1.27-1.33], 1.17 in January (95% CI: 1.14-1.20), and 1.16 in March (95% CI: 1.13-1.29). These peaks in search volumes occur within the same months as those observed in the corresponding police calls categorised as domestic violence. Police data also showed somewhat higher volumes in April. Conclusions: The study suggests that Google search volumes can be used to study the highest peaks in seeking help for domestic violence in countries with a high level of Internet usage and no available police data.
Authors: Corinne A Riddell; Krista Neumann; N Jeanie Santaularia; Kriszta Farkas; Jennifer Ahern; Susan M Mason Journal: J Med Internet Res Date: 2022-06-13 Impact factor: 7.076
Authors: Corinne A Riddell; Kriszta Farkas; Krista Neumann; N Jeanie Santaularia; Jennifer Ahern; Susan M Mason Journal: Prev Med Date: 2022-08-23 Impact factor: 4.637