Research That Matters (January 17 - 20, 2008)

We are proposing this workshop to provide a learning experience to social work researchers, practitioners, and students on the construction and psychometric validation of a culturally-valid acculturation scale.

Population change is one constant dynamic in world history. Today, more than 150 million migrants have crossed national boundaries in the world. The United States receives less than 2% of the world's migrants on an annual basis. The foreign-born population of the US is 9.5 percent of the total population (in 2000). This can be compared to the 2000's proportions of 22.7 in Australia; 16 percent in Canada; 6.3 in France; 7.3 in Germany; 3.9 percent in Great Britain; and 5.7 in Sweden. In 2001, 250,346 immigrants resettled in Canada. In total, immigrants account for 17 per cent of Canada's population. Similarly, net migration into the UK has averaged 166,000 a year over the last 7 years. Acculturation is an important social and psychological construct, because it has been confirmed as a determinant or moderator of health and mental health outcomes among immigrants invariant of racial or ethnic backgrounds. Acculturation involves changes at both the group level and the individual psychological level resulting from sustained contact between two or more distinct cultures (Berry, Trimble, & Olmedo, 1986).

Design:

We will present three stages of scale construction and discuss the theoretical and methodological considerations in each stage.

Stage 1: Generate the universe of acculturation items

a. Interviewed nine immigrants (equal country sample) to obtain ethnographic information on their perceived acculturation. b. Researchers generated independently an exhaustive list of behavioral items with Asian immigrant groups in US, UK, and Canada. c. Items were pooled to create a master list of item pools.

Stage 2: Create a self-reported acculturation questionnaire

a. Selected the appropriate response format. b. Created an expert panel to eliminate items not sensitive to the research construct. Items showing low inter-rater reliability less than .90 kappa coefficient were eliminated. c. The remaining items were retained as the acculturation questionnaire.

Stage 3: Validation studies

Study 1: Total sample of 300 adults (aged 18 or more) were recruited in equal proportion in the respective ethnic groups.

1. We will report the data analysis to determine the initial psychometric properties:

a. Descriptive statistics; b. Frequencies to determine low-occurrence items, which could be dropped later. c. Examine possible group differences. d. Conduct factor analysis to reduce items based on covariation.

Study 2: A second sample of 300 adults (aged 18 or more) was recruited in equal proportion in the respective ethnic groups.

1. We will report the second validation study based on the reduced scale in Study 1. Statistics will be presented as follows:

a. Item-total correlation matrices. b. Assess group differences in factor scores. c. Confirmatory factor analysis (CFA) with maximum-likelihood method was used to test the models to determine the goodness-of-fit to the data.

Conclusion and Implication

The data confirm a multi-stage development of a culturally valid acculturation scale from the ground up. Future validation research will be discussed.