Authors:
Luciane C. Araújo
1
;
Marco A. Casanova
2
;
Luiz André P. P. Leme
3
and
Antonio L. Furtado
2
Affiliations:
1
Brazilian Institute of Geography and Statistics, Rio de Janeiro, Brazil
;
2
Dept. Informatics, PUC-Rio, Rua Marquês de São Vicente, 225, Rio de Janeiro, Brazil
;
3
Institute of Computing, Federal Fluminense University, Niterói, Brazil
Keyword(s):
Survey Questionnaires, Domain-specific Languages, Statistical Surveys.
Abstract:
The use of surveys permeates the economy, ranging from customer satisfaction measurement to tracking global economic trends. At the core of the survey process lies the codification of questionnaires, which vary from simple lists of questions to complex forms that include validations, computation of derived data, use of triggers to guarantee consistency, and dynamic creation of objects of interest. Questionnaire specification is part of what is called survey metadata and is a key factor for the quality of the data collected and of the survey itself. In this context, the paper first introduces a comprehensive complex questionnaire model. Then, based on the model, it proposes a prototype domain-specific language (DSL) for modeling complex questionnaires, called SLang. The paper also describes a prototype implementation of SLang and an evaluation of the usefulness of the language in practical scenarios.