Pedagogy of the Logic Model: Teaching Undergraduates to Work Together to Change Their Communities

Undergraduate community psychology courses can empower students to address challenging problems in their local communities. Creating a logic model is an experiential way to learn course concepts by “doing.” Throughout the semester, students work with peers to define a problem, develop an intervention, and plan an evaluation focused on an issue of concern to them. This report provides an overview of how to organize a community psychology course around the creation of a logic model in order for students to develop this applied skill. Two undergraduate student authors report on their experience with the logic model assignment, describing the community problem they chose to address, what they learned from the assignment, what they found challenging, and what they are doing now in their communities based on what they learned.     

Two-Year Reliability of the Propensity for Abusiveness Scale

The Propensity for Abusiveness Scale (PAS; Dutton [1995a] J. Fam. Violence 10[2]: 203–221) has been shown to correlate with both physical and psychological abuse in a variety of samples including gay men, assaultive males, and male and female college students (Clift, 2001; Dutton et al., 2001). For the current study, 27 male and 37 female university students completed the PAS during the 1998–1999 and the 2000–2001 school years. A 2-year test–retest reliability coefficient of r = .851 was found for women and a reliability coefficient of r = .629 was found for men. The combined sample had a reliability coefficient of r = .774. Moderate test–retest reliability coefficients were also found for each of the subscales.     

A comparison of adjustment methods to test the robustness of an STR DNA database comprised of 24 European populations

An aim of the European Network of Forensic Science Institutes (ENFSI) is to produce a DNA database of second generation multiplex (SGM) STR profiles that is representative of the resident cosmopolitan populations. To achieve this, data were collected from 24 different populations. All of the data were combined to form one database of 5700 profiles from which allele proportions were calculated. The robustness of this combined European database was tested by estimating parameter d for every DNA profile, where d=log(10)(Pm(c)/Pm(E)) Pm(c) is the match probability of the profile calculated from its cognate database and Pm(E) is the match probability of the combined European database. Overall there was a small tendency for Pm(c)>Pm(E) primarily because of sampling bias. This bias was removed by the simple expediency of applying an adjustment factor to the calculation of Pm(E). These were selected from the Balding size bias correction, the Balding and Nichols Fst correction, a minimum allele proportion (between 0.01 and 0.02), an upper bound of a 95% confidence interval (CI) and a lower bound on the genotype match probability. It was demonstrated that a single European database is a feasible proposition. A combination of different adjustment methods can be used to ensure that the result is conservative relative to the cognate database, and their effect measured by parameter d.