Hispanics are the largest and fastest growing minority group in the nation (U. S. Census Bureau, 2004). It is estimated that by year 2020 there will be almost 60 million Hispanics, totaling approximately 18% of the population. As Extension professionals, we continually develop and adapt programs to meet the changing needs of clientele, and there are numerous articles highlighting these efforts with Hispanic audiences (e.g., Anding, Fletcher, Van Laanen, & Supak, 2001; Israelsen, Young, & Boman, 2006; Kock, 2003; Peterson et al., 2008). An important part of this work is evaluating whether programs are successful in obtaining desired outcomes and impacts; however, most of the available assessment tools have been developed and validated with English-speaking, middle class samples, resulting in measures that may not generalize to other populations. Thus, Extension professionals often either translate existing scales or develop new questionnaires to assess program outcomes and impacts.
Research on the development and cultural adaptation of surveys for use with Hispanic audiences has generally focused on the translation and meaningfulness of the individual items and process undertaken (e.g., Brislin, 1970; Carroll, Holman, Segura-Bartholomew, Bird, & Busby, 2001; Cha, Kim, & Erlen, 2007; Farner, Cutz, Farner, Seibold, & Abuchar, 2006; Sperber, Devellis, & Boehlecke, 1994; Warrix, Nieto, & Nicolay, 2006). Few studies have focused on the cultural adaptation and translation of the rating scales or response categories (e.g., Arce-Ferrer, 2006b; Clarke, 2000; Hui & Triandis, 1989) and there are no published studies in the Journal of Extension on this topic.
One might ask why consideration of the rating scales is important. A rating scale is a set of categories that elicit information about the direction and strength of a participant's feelings, beliefs, opinions, and knowledge. Therefore, it is important to ensure appropriate practices are followed in the development, cultural adaptation, and translation of rating scales because the validity of the data, inferences about program outcomes and impacts, as well as cross-cultural comparisons depend, in large part, on the rating scale. Thus, while previous research has clearly outlined some of the best practices and methodological issues with regard to the translation of measures, the importance of rating scales warrant greater attention in both research and practice.
In this article, research on extreme responding and techniques for reducing this response style are reviewed and recommendations for Extension professionals outlined. The focus is on Spanish-speaking, Hispanic populations of Mexican origin. This population was chosen for two reasons. First, Mexican Americans currently comprise 66% of all Hispanics in the United States (U.S. Census Bureau, 2009). Second, there is some evidence that findings may differ among other Hispanic populations from Central America, the Caribbean, and South America (e.g., Gibbons, Zellner, & Rudek, 1999). Whether the goal is to adapt and translate an existing questionnaire for cross-cultural use, develop a new assessment tool, or interpret cross-cultural findings from one's own or others' research, this article provides a guide for thinking about the forgotten half of evaluation instruments—the rating scales. Inattention to these issues could lead to misleading results and interpretations.
Dogan, S. J., Sitnick, S., & Ontai, L.
2012,
The forgotten half of program evaluation: A focus on the translation of rating scales for use with Hispanic populations.,
Journal of Extension 50(1) Article 1FEA5.
