Utilizing Factor Analysis to Understand the Needs of Asian American Students

According to the 2000 U.S. Census, there were over 12 million Asian Americans residing in the U.S., representing an increase of 50% since 1990 (U.S. Census Bureau, 2002). Over the course of a similar time span, there has been a dramatic increase of Asian American college students. Wilds (2000) reported that the enrollment of Asian Americans increased 73% from 1988 to 1997. In 1997, Asian Americans comprised 4% of the U.S. population and 6% of the total enrollment in higher education. Educational research regarding Asian Americans, however, has not been commensurate with their growth. Hune and Chan (1997) have argued that Asian Americans are not included in research because they are considered a "model minority." First coined in 1966 by William Peterson, "the model minority" suggests that Asian Americans are an example of the modern day American success story. The idea follows that Asian Americans have overcome social problems. Scholars have argued that the model minority myth persists (Suzuki, 2002) and that it has served to obscure the social, economic, educational, and psychological concerns that Asian Americans experience (Hune & Chan, 1997). Liang and Sedlacek (in press) found some indication that student affairs practitioners subscribe to the model minority myth.

In recent years, there has been a call to guide practice through research (Komives, 1999). Campus units have sought to address the concerns of Asian American students by conducting focus groups and/or distributing surveys; Majors and Sedlacek (2001), however, suggested that multiple surveys by different units on a campus may obscure important dimensions underlying student issues, and that the use of factor analysis is one way to avoid this problem. Applying factor analysis to large surveys can help identify key factors that underlie information provided by students. Factor analysis serves to reduce a large number of student responses in surveys to fewer understandable constructs. These constructs serve as a way to think about the most important aspects of the data, making it easier for administrators to develop more appropriate programs. Finally, disaggregating student responses by racial or ethnic groups provides data that will be more reflective of each population. Multicultural scholars have long argued that experiences of racial groups will differ one from another and that it is important not to lose sight of between-group differences. The purpose of this study was to use factor analysis to understand and organize service delivery for Asian American college students.

Method

Participants

Participants were 417 (47% female) first-year Asian American students from a large mid-Atlantic state university who attended a summer orientation program. Asian American students represented 13.8% of the 3,021 students who completed the instrument. The median age of the Asian American students was 18 years.

Instrument

The University New Student Census (UNSC) is an on-line survey designed to elicit responses regarding student perceptions, attitudes, expectations, and interests. Students attending the summer orientation program completed the survey on line in computer labs. Students who attended the summer orientation program represented 90% of all incoming students at the university. More than 90% of those students who attended orientation completed the survey. For the purposes of this study, however, only responses of incoming Asian American students attending summer orientation were analyzed. Individuals from student and academic affairs units, committees, and researchers generated the survey items that would help them serve incoming students. Test-retest reliability was estimated at .85 for scores from a sample of students. Fifty-two items, using a 5-point Likert-type scale ranging from 1 (strongly agree) to 5 (strongly disagree), that measured a variety of issues, were used in this study.

Results

In order to identify the underlying constructs, the 52-items were submitted to a principal component factor analysis, with Varimax rotation that accounted for 62% of the common variance among the items. Communalities were estimated...