Welcome to the home page for the multiple informant & multiple outcome data project. Multiple informant (or multiple source) data are often collected in surveys designed to measure mental health outcomes and risk factors in community or service-based samples (though they arise in other settings as well). For example, in assessments of a child's mental health status, parents, teachers, children, clinicians, peers, and trained observers have served as informants or sources. Multiple sources have also become increasingly common in hospital-based and outpatient-based assessments of treatments for mental illness and service utilization. Contemporary studies commonly use two or three informants as data sources.

An inherent feature of multiple informant data is that one anticipates discordant reports. If there is no discordance, the additional reports provide no new information. Multiple sources are used to provide data either on independent variables (risk factors or predictors), or dependent variables (outcomes), or both.

Similar issues arise in the use of multiple outcome data, which are often collected when analysis is to consider a range of responses. In the case of randomized clinical trials, there may not be a clear concensus on the most important clinical outcome. Or, a single measure may not be enough to characterize the effect of a treatment. In such situations, it is necessary to consider several outcomes in a statistically appropriate manner.

The project develops methodology for the analysis of multiple informant and multiple outcome data. Take a look at what's new. Other information includes background information on multiple informants and multiple outcomes, frequently asked questions, macros and examples, publications, and a list of project members.

This research is funded by NIMH grant R01-MH54693.

Last modified on January 16, 2012.