Experimental and Quasi-Experimental Designs in Visitor Studies: A Critical Reflection on Three Projects

Identifying causal relationships is an important aspect of research and evaluation in visitor studies, such as making claims about the learning outcomes of a program or exhibit. Experimental and quasi-experimental approaches are powerful tools for addressing these causal questions. However, these designs are arguably underused in visitor studies. In this article, we offer examples of the use of experimental and quasi-experimental designs in science museums to aide investigators interested in expanding their methods toolkit and increasing their ability to make strong causal claims about programmatic experiences or relationships among variables. Using three designs from recent research (fully randomized experiment, posttest only quasi-experimental design with comparison condition, and posttest with independent pretest design), we discuss challenges and tradeoffs related to feasibility, participant experience, alignment with research questions, and internal and external validity. We end the article with broader reflections on the role of experimental and quasi-experimental designs in visitor studies.     

Social networks: A link between psychiatric epidemiology and community mental health

This paper examines the concept of social network as a mediating construct linking psychiatric epidemiology and community mental health. A selective review of studies investigating the structural and interactional features of the social networks of psychiatrically impaired persons is presented and the implications of their results are discussed. It is concluded that social networks provide a unifying framework potentially capable of linking empirical research with clinical practice.This work is supported in part by grant #AA03506-04 from the National Institute on Alcohol Abuse and Alcoholism.     

A factorization procedure for finite algebras

A procedure is developed for decomposing any finite algebra into a minimal set of maximally independent simple homomorphic images, or factors, of the algebra. The definition of admissible sets of factors is made in relation to the congruence lattice of the algebra, and generalises the notion of an irredundant reduction in a modular lattice. An algorithm for determining all possible sets of factors of a given finite algebra is derived and an index for measuring the degree of independence of factors is defined. Applications of the technique to finite algebraic models within the social psychological domain are presented and include factorizations for certain semigroups of binary relations and for a class of finite semilattices.     

Testing the Impact of a Computer Guide on Visitor Learning Behaviors at an Interactive Exhibit

ABSTRACT

There is a recognized need to rigorously examine the efficacy of approaches to supporting informal learning. In this study, we used a 2 × 2 factorial experimental design to test the impact of a computer guide on 3 proximal measures of visitor learning at an interactive math exhibit. In total, 128 families were systematically assigned to engage with the exhibit either with or without access to a supplementary computer kiosk. Visitor groups with access to the computer spent longer, on average, at the exhibit and engaged in more mathematical behaviors compared to other groups. However, based on interviews, visitors with access to the computer were less likely to fully articulate the mathematical relations in the exhibit. These results suggest that although computer guides are a promising approach to supporting visitor engagement, they may, unless carefully designed, undermine other learning outcomes.     

Attributional style in psychosis--the role of affect and belief type

People holding persecutory beliefs have been hypothesised to show a self-serving attributional style, which functions to protect self-esteem Bentall, Corcoran, Howard, Blackwood, and Kinderman (2001). Experimental support for this has been mixed. Freeman et al. (1998) suggested depressed and grandiose subgroups of those with persecutory beliefs might explain events differently. In this study, 71 participants completed measures of delusional beliefs, depression and attributional style. We hypothesised that those with persecutory beliefs would form grandiose and depressed subgroups, and that a self-serving attributional style would characterise only the grandiose subgroup. Hypotheses were partially confirmed. Clear subgroups were evident and only those with both persecutory and grandiose beliefs showed an externalising attributional style for negative events. Depression, irrespective of co-occurring persecutory beliefs, was related to a reduced self-serving bias and an externalising attributional style for positive events. On their own, persecutory beliefs were not related to any particular attributional style. Depressed and grandiose subgroups of those with persecutory beliefs might account for some of the inconsistencies in the attribution literature. Even within a single symptom group, care should be taken in both research and therapy to consider individual symptom patterns.