(Weiss, J. W., & Weiss, D. J. (2002).  Recruiting Asian-American adolescents for behavioral surveys.  Journal of Child and Family Studies, 11, 143-149.)

Recruiting Asian-American Adolescents for Behavioral Surveys

Jie Wu Weiss

University of Southern California

David J. Weiss

California State University, Los Angeles

            Asian societies have distinct languages, cultural traditions, and histories.  An important empirical question is whether the various ancestral countries yield immigrants who differ in psychologically interesting ways.  For any research issue inspiring speculation that cultural background is important, we can try to examine whether differences across subgroups are large relative to differences among individuals within the same subgroup.  In order to explore such issues, researchers must be able to recruit sufficient numbers of representatives of each subgroup.  We discuss lessons learned while carrying out a study that sought several hundred adolescents from four East Asian subgroups (Weiss, 2001).

Locating Potential Participants

Many Asian-Americans live in ethnic communities.  Within the Los Angeles area, there are street signs identifying Chinatown, Koreatown, Little Saigon, and Little Tokyo.  Residential association occurs for various reasons, such as speaking a familiar language in the neighborhood, easy access to traditional foods and cultural events, and recruitment by friends and relatives.  Not only is it more efficient to recruit targeted participants in these particular areas, but also residents may have closer ties to their culture and therefore be more representative.  However, not all Asian-Americans live in such enclaves.  Those who are more acculturated may be more scattered.   Therefore, exclusive reliance upon recruitment within the ethnic community may yield a sample that is biased with respect to factors associated with acculturation. 

Generally, adolescent participants will be recruited through schools (Riesch, Tosi, & Thurston, 1999).  There is little practical alternative to accessing adolescents through schools, although it is known that at least for some issues, school dropouts behave differently from attenders (Swaim, 1997).  A pragmatic issue is the search for schools whose students are drawn from the subgroups sought for the study.  Beyond simply seeking out the schools in neighborhoods known to contain the desired population, we have found the Internet useful.  School districts commonly display the ethnic composition of their student bodies. 

Issues Of Acculturation

The category of Asian-American encompasses descendants of immigrants who came to the United States in the 1850s as well as immigrants who arrived yesterday.  Degree of acculturation can be a powerful predictor of behavior (Florsheim, 1997; Wong, 1999), so for many issues it is useful to gather information about this variable.

Two aspects of acculturation might be expected to affect the behavior of Asian American immigrants with respect to research participation.  The length of time the individual has spent in the United States is important, although it may be tempered by the degree of contact with people and institutions outside the ancestral culture.  Because adolescents go to school, they inevitably are exposed to American culture and the English language, whereas their parents may be able to survive within an enclave that is not very different from the home country.  This intergenerational difference is relevant to gaining consent from parents to allow their children to serve as participants in research projects.  If the parents of potential participants are not fluent in English, then it will be necessary to provide information and consent forms in their own languages.

For recent immigrants, the extent of differences between the ancestral culture and that of the United States also may play a role in cooperation.  While Americans are generally familiar with and appreciative of research, enriched explanation and justification of the project within the informational materials provided to parents may be advisable.  Associating the project with the prestige of a university may be valuable, especially with respect to immigrant parents (Berg, 1999).

Securing Cooperation From Schools

The process of recruitment via schools generates an array of seldom-discussed practical challenges.  Our experience with the recruiting effort (Weiss, 2001) has provided valuable insights that are relevant to any study in which intermediaries control access to participants.

            There is an administrative chain of command to be negotiated.  At the investigator’s home institution, the Institutional Review Board (IRB) must be convinced of the value of the study.  IRBs tend to be especially cautious when children are involved (Rogers, Schwartz, Weissman, & English, 1999).  Items exploring sexual behaviors or drug usage raise concerns.  The risk/benefit ratio of the research must be clarified to the IRB’s satisfaction.  One should not think of the IRB merely as a hurdle, however.  Their approval carries considerable weight with the educational administrators whose support is needed.  Additionally, IRB members may have useful contacts with community representatives.

            School districts have a hierarchical structure.  At the top is the district (or sub-district) superintendent, whose approval is critical.  Weiss (2001) was able to contact four superintendents regarding the study, by mail and phone.  Two agreed to cooperate, and two refused.  In two other districts, the decision was left to principals.  Refusal by the superintendent is the end of the line for that district, but assent does not guarantee access to participants.  The school principal also controls access, whether at the high school, junior high, or middle school level.  The most successful strategy has included an initial brief telephone conversation asking for face-to-face contact.  Prior to the meeting, the researcher sends a letter to the principal describing the study and the specific goals within the particular school.  Documentation of approval by the IRB and other agencies, along with citations of additional schools that have already agreed to cooperate, lends credibility to the project.  Similarly, if the research has received financial support, the prestige attaching to funding may help to convince the principal that the study is worthwhile. 

Most principals are favorably disposed toward research.  However, they are very busy.  It is the researcher’s responsibility to demonstrate that cooperation requires minimal effort on the part of the principal.  If a principal declines to cooperate, there is no recourse.  A positive response may be accompanied by a request for the researcher to carry out subsequent negotiations with a designated member of the staff, usually a health coordinator or vice principal.  Cementing the scientific connection by promising to provide results at the conclusion of the study is not only a courtesy, but will promote a long-term relationship that may be useful in future work.  Ten of the 13 principals approached by Weiss (2001) agreed to cooperate.

            Once the principal decided to allow access, teacher cooperation was universal.  Teachers must get the students to bring in consent forms signed by parents and set aside classroom time for administering questionnaires.  Some teachers prefer to handle the questionnaires themselves, while others permit a member of the research team to pass them out and collect the responses.  It is vital that instructions to participants be written clearly and comprehensively, because the researcher cannot count on having an opportunity to address questions that participants may have.

            A possible reason that we have been able to get so much cooperation from school personnel is our routine reliance on extrinsic rewards.  The inducements we offer are too small to be viewed as bribes, but still this issue must be handled with delicacy.  We have offered token payments at the time when school officials expressed their intention to cooperate with data collection, conveying our gratitude for their understanding and support of our research project.  The rewards have not been given in cash, as that might be seen as inappropriate.  The usual pattern was to give an art product, such as a small plate or scarf, worth about $10-$15, during the first meeting with the principal.  The second meeting was arranged by the principal, who identified the teachers of the classes containing students within the ages sought.  At that second meeting, the principal usually received an envelope containing a $50 gift certificate from a chain store; each teacher received an envelope containing a $30 gift certificate.  At that time, the teachers also were given the research materials.  Later, after the data were collected, teachers were given an art product as a parting gift.

Securing Cooperation From Parents

For most studies involving adolescents, parental consent must be affirmed.  This entails providing written information about the study, as well as an explanation of the right to refuse to allow participation.  The researcher needs to be sensitive to cultural issues in preparing these materials.  Asian-Americans are in general not eager to avail themselves of psychological services (Sue, Fujino, Hu, Takeuchi, & Zane, 1991), and psychology as a field is viewed with skepticism (Atkinson, Ponterotto, & Sanchez, 1984; Tsui & Schultz, 1985).  On the other hand, there is respect for medicine, so casting the study as one regarding health issues may enhance participation rates.  At the same time, one must be careful to explain that there are no potential health risks, as there might be in a medical study.

Parents may be concerned that the child’s responses will have adverse consequences.  The study may examine illegal acts, such as drug use, or difficulties in family functioning, that parents might not like to reveal.  It is therefore very important to emphasize the privacy guarantees regarding the data.  Anonymity of responses is the most effective of these guarantees (Ong & Weiss, 2000), but may require careful explanation to help parents to distinguish signed consent forms from anonymous response forms.  The consent form should also explain that the researcher has no interest in individual reports, and that the data will be analyzed and presented only on a group basis.  It may help to illustrate positive societal contributions made by comparable studies.

            It may be tempting to rely upon passive procedures to gain parental consent (Severson & Ary, 1983), if the IRB permits this approach.  Passive consent means that unless there is evidence that the parent has explicitly refused to allow participation, the decision rests solely with the potential participant.  Passive consent procedures circumvent parental apathy (Severson & Biglan, 1989), and have the pragmatic advantage that they do not require the student messenger to bring the consent form home, get it signed, and bring it back in order to permit participation.  Active consent from parents can also be obtained by mail or phone (Moberg & Piper, 1990); however, such efforts can be costly (Ellickson & Hawes, 1989).  Passive procedures yield substantially higher participation rates (Kearney, Hopkins, Mauss, & Weisheit, 1983).  It has also been suggested that they provide samples that are more representative as well (Anderman et al., 1995; Dent et al., 1993), probably because the less compliant children are not as likely to return the consent forms.

            On the other hand, active consent requirements are more honest, in that the researcher can be certain the parent of every participant has seen the consent form and truly given approval.  Active consent avoids the potential problem that parents who were not informed about the study subsequently raise the kind of objections that make IRBs very nervous (Jason, Pokorny, & Katz, 2001; Mammel & Kaplan, 1995).

            The IRB charged with oversight of the Weiss (2001) study considered the researcher’s application to use passive consent procedures, but did not grant the request.  Using students as messengers, signed parental approval was obtained for some 65% of the approximately 4500 class members who received the consent form.  Unfortunately, we do not know how many of the forms failed to reach parents.  Ideally, non-consent reflects a deliberate choice, perhaps occasioned by reluctance to have the child report on what the parent thought the study would be asking about (Cochran, Mosteller, & Tukey, 1953).  We view participation rate as an outcome that should be reported routinely.

Securing Cooperation From Participants

            We have found it useful to emphasize that the study provides an opportunity for the participant’s voice to be heard.  Especially for Asian-American students, whose parents often expect children to accede to their opinions, this sense of empowerment serves to strengthen the appeal of participation.  The adolescent must also be assured that divulging past behaviors and attitudes will not have adverse consequences, and that neither parents nor teachers will be able to attribute responses to individuals.  As the questionnaires are distributed, we announce that either chocolate bars or snack bars will be available upon completion.  Participation is spurred by both incentives and the opportunity to express one’s opinion (Cauce, Ryan, & Grove, 1998; Groves, Singer, & Corning, 2000).  None of Weiss’s (2001) students who returned a consent form refused to complete the questionnaire.

            Usually, a class will contain students who are not members of the target group.  It is generally a good idea to attempt to recruit the entire class.  That strategy avoids possible paranoia on the part of the students who may feel singled out, and is often logistically easier.  Filtering can be carried out during the data-analytic phase of the project.  In addition, it may turn out that someday, the data sitting in the archives will be just what is needed to answer a question that had not been considered at the time of the study.

The Research Team

            If a study aims for responses from members of several subgroups, it may be useful to enlist collaborators with appropriate language fluency.  Parents whose English is weak may give consent without full understanding, but the consent will not be informed even though a signature is sufficient to satisfy legal requirements. 

            We do not know whether having a person with an Asian face administering the materials makes a difference.  Matching the ethnicity of the researcher and the participant has proven valuable for African-Americans in clinical studies (Thompson, Neighbors, Munday, & Jackson, 1996).  Issues of trusting the researcher have been raised for other ethnic groups (Levkoff, Levy, & Weitzman, 2000; Stoddart et al., 2000), but we are unaware of data bearing on this question for Asian-Americans.  Similarly, historical antipathies among subgroups, e.g., Chinese vs. Japanese (Deng, 1997), may affect participation and veracity.  

Conclusion

            Although every researcher who ventures beyond the introductory psychology subject pool must face the issue, there are few published guidelines for effective recruitment.  It might be worthwhile to systematically investigate whether the features we have recommended, especially those involving financial costs, really do have an impact.  We know that sampling biases can affect research conclusions; we know little regarding the ways in which our recruiting procedures can affect the integrity of our samples.

ACKNOWLEDGEMENTS

            Jie Wu Weiss was partially supported by a grant from the State of California’s Tobacco-Related Disease Research Program (9DT-0090).  We wish to thank Jean S. Phinney for a critical reading of the manuscript.

References

Anderman, C., Cheadle, A., Curry, S., Diehr, P., Shultz, L., & Wagner, E. (1995).  Selection bias related to parental consent in school-based survey research.  Evaluation Review, 19, 663-674.

Atkinson, D. R., Ponterotto, J. G., & Sanchez, A. R. (1984).  Attitudes of Vietnamese and Anglo-American students toward counseling.  Journal of College Student Personnel, 25, 448-452.

Berg, J. A. (1999).  Gaining access to underresearched populations in women's health research.  Health Care for Women International, 20, 237-243.

Cauce, A. M., Ryan, K. D., & Grove, K. (1998).  Children and adolescents of color, where are you?   Participation, selection, recruitment, and retention in developmental research.  In V. C. McLoyd & L. Steinberg (Eds.), Studying minority adolescents: Conceptual, methodological, and theoretical issues (pp. 147-166).  Mahwah, NJ: Erlbaum.

Cochran, W. G., Mosteller, F., & Tukey, J. W. (1953).  Statistical problems of the Kinsey report.  Journal of the American Statistical Association, 48, 673-716.

Deng, Y. (1997).  Chinese relations with Japan: Implications for Asia-Pacific regionalism.  Pacific Affairs, 70, 373-391.

Dent, C. W., Galaif, J., Sussman, S., Stacy, A., Burton, D., & Flay, B. R. (1993).  Demographic, psychosocial, and behavioral differences in samples of actively and passively consented adolescents.  Addictive Behaviors, 18, 51-56.

Ellickson, P. L., & Hawes, J. A. (1989).  An assessment of active versus passive methods for obtaining parental consent.  Evaluation Review, 13, 45-55.

Florsheim, P. (1997).  Chinese adolescent immigration factors related to psychosocial adjustment.  Journal of Youth and Adolescence, 26, 143-163.

Groves, R. M., Singer, E., & Corning, A. (2000).  Leverage-salience theory of survey participation: Description and an illustration.  Public Opinion Quarterly, 64, 299-308.

Jason, L. A., Pokorny, S., & Katz, R. (2001).  Passive versus active consent: A case study in school settings.  Journal of Community Psychology, 29, 53-68.

Kearney, K. A., Hopkins, R. H., Mauss, A. L., & Weisheit, R. A. (1983).  Sample bias resulting from a requirement for written parental consent.  Public Opinion Quarterly, 47, 96-102.

Levkoff, S. E., Levy, B. R., & Weitzman, P. F. (2000).  The matching model of recruitment.  Journal of Mental Health and Aging, 6, 29-38.

Mammel, K. A., & Kaplan, D. W. (1995).  Research consent by adolescent minors and institutional review boards.  Journal of Adolescent Health, 17, 323-330.

Moberg, D. P., & Piper, D. L. (1990).  Obtaining active parental consent via telephone in adolescent substance abuse prevention research.  Evaluation Review, 14, 315-323.

Ong, A. D., & Weiss, D. J. (2000).  The impact of anonymity on responses to “sensitive” questions.  Journal of Applied Social Psychology, 30, 1691-1708.

Riesch, S. K., Tosi, C. B., & Thurston, C. A. (1999).  Accessing young adolescents and their families for research.  Image: Journal of Nursing Scholarship, 31, 323-326.

Rogers, A. S., Schwartz, D. F., Weissman, G., & English, A. (1999).  A case study in adolescent participation in clinical research: Eleven clinical sites, one common protocol and eleven IRBs.  IRB: A Review of Human Subjects Research, 21, 6-12.

Severson, H. H., & Ary, D. V. (1983).  Sampling bias due to consent procedures with adolescents.  Addictive Behaviors, 8, 433-437.

Stoddart, M. L., Jarvis, B., Blake, B., Fabsitz, R. R., Howar, B. V., Lee, E. T., & Welty, T. K. (2000).  Recruitment of American Indians in epidemiologic research: The strong heart study.  American Indian & Alaska Native Mental Health Research, 9, 20-37.

Sue, S., Fujino, D., Hu, L., Takeuchi, D., & Zane, N. (1991).  Community mental health services for ethnic minority groups: A test of the cultural responsiveness hypothesis.  Journal of Consulting and Clinical Psychology, 59, 533-540.

Swaim, R. C. (1997).  The effect of school dropout rates on estimates of adolescent substance use among three racial/ethnic groups.  American Journal of Public Health, 87, 51-55.

Thompson, E. E., Neighbors, H. W., Munday, C., & Jackson, J. S.  (1996).  Recruitment and retention of African-American patients for clinical research: An exploration of response rates in an urban psychiatric hospital.  Journal of Consulting and Clinical Psychology, 64, 861-867.

Tsui, P., & Schultz, G. W. (1985).  Failure of rapport: When psychotherapeutic engagement fails in the treatment of Asian clients.  American Journal of Orthopsychiatry, 55, 561-569.

Weiss, J. W. (2001, December).  Self-image, family, and acculturation: Why Asian-American adolescents smoke.  Poster session presented at the annual investigator meeting of the Tobacco-Related Disease Research Program, Los Angeles.

Wong, S. K. (1999).  Acculturation, peer relations, and delinquent behavior of Chinese-Canadian youth.  Adolescence, 34, 107-119.