Canadian Problem Gambling Index

Developing the CPGI

The CPGI was launched in 2001 as the first project of the collaborating group of funders that has now become the CCGR. Since its launch, the CPGI has become internationally recognized as a robust measure of gambling behaviour. The CPGI includes the nin-item Problem Gambling Severity Index, which has been used extensively across Canada and in Australia, Great Britain, Iceland and Norway to measure the prevalence of problem gambling.

• RESEARCH TEAM: Jackie Ferris and Harold Wynne
• STATUS: Complete (2001)

Evaluating the CPGI's performance

As a result of its widespread use, researchers gained significant experience and insight in the use of the CPGI. In 2006, the CCGR funded a study to capture this experience and make recommendations for the further development and enhancement of the CPGI. To complement these findings, the CCGR then funded a 2007 review that examined critiques of the CPGI and secondary analyses emanating from existing surveys. Combined, these projects led to a plan for a series of prioritized CPGI enhancement initiatives to ensure the CPGI remains a state-of-the-art instrument for conducting population surveys on the nature, extent and consequences of problem gambling among adults.

• RESEARCH TEAM: John McCready and Edward Adlaf
• STATUS: Complete (2006)

Refining the CPGI

In 2009, the CCGR began funding this series of prioritized research initiatives, all of which were overseen by an expert advisory panel to ensure rigorous scientific methodology throughout the enhancement process. The expert advisory panel included Norbert Schmitz (McGill University), Sherry Stewart (Dalhousie University), and Joël Tremblay (Université du Québec à Trois-Rivières).

PROJECT: Improving the Psychometric Properties of the Problem Gambling Severity Index

• RESEARCH TEAM: Shawn Currie, David Casey and David Hodgins
• STATUS: Complete (2010)

The first initiative, an analysis of the CPGI's psychometric properties and the cut-off points for gambling risk levels, was completed in 2010. The project analyzed the performance of the CPGI's Problem Gambling Severity Index (PGSI) in three large population datasets and examined the viability of alternative scoring and weighting schemes. Results show that two of the PGSI's subtypes - the non-problem and problem gambling subtypes - represent distinct and well-defined groups, but that the original classification of low-risk and moderate-risk subtypes fits less well. As a result, the PGSI scoring cut-off rules were modified for the low-risk and moderate-risk subtypes. The CCGR encourages researchers to report all four subtypes separately, as they represent meaningfully different groups of gamblers.

PROJECT: Developing a CPGI Population Harm Supplement

• RESEARCH TEAM: Michael Bagby, Lena Quilty and Chris Watson
• STATUS: Complete (2011)

The second initiative consisted of the development of new CPGI items to more broadly assess the harm associated with gambling in survey populations.

Project Resources