Campbell, J. R., Tirri, K., Ruohotie, P., & Walberg, H. (Eds.). (2004). Cross-cultural Research: Basic Issues, Dilemmas and Strategies.

Campbell, J. R., Tirri, K., Ruohotie, P., & Walberg, H. (Eds.). (2004). Cross-cultural Research: Basic Issues, Dilemmas and Strategies. Hameenlinna: Research Center for Vocational Education (194 pp., $45.00 cloth, ISBN: 951-44-5940-7).

Cross-cultural Research: Basic Issues, Dilemmas, and Strategies is an innovative primer for conducting international research. This logical and well-organized text is based upon concrete examples from the ongoing (1972 to present) Olympians Research Project (ORP) which explores how talent is developed in Germany, Finland, Taiwan, China, and the United States. Prominent scholars from Germany, Finland, and the United States disclose common issues in conducting accurate cross-cultural studies, offer applications and solutions to common quandaries, cite the weaknesses of prior large scale international studies, tout the latest Bayesian statistical methods as a possible panacea, and provide strategies for creating cross-cultural instruments that represent the attributes of diverse cultures.

The editors also focus upon the pressing need to conduct international studies with valid cross-cultural instruments as accurate information promotes cultural understanding and economic productivity. Citing the heightened rise of ethnic minority populations within most countries over the last four decades, specifically within the United States and Europe, the editors discuss the logic behind harnessing the attributes of diverse cultures as understanding eradicates ethnic bigotry/ethnic cleansing, and promotes global economic stability.

This user-friendly text is a must read for educational and government policy makers, cross-cultural researchers, and the research community at large. As educational and government policies are influenced by the findings of large-scale studies measuring student achievement, the authors convincingly and substantively compel the community to re-evaluate the constructs used in previous large-scale international studies. Principal investigators will also benefit from the ORP researchers' caveats and easy to apply solutions, examples from the International Academic Olympiad Studies, the latest Bayesian statistics, and pragmatic blueprints for creating valid cross-cultural instruments. This all inclusive primer is a pioneer in the research field.

Cross-cultural Research: Basic Issues, Dilemmas, and Strategies consists of 10 chapters and is logically organized. It is current, well-researched, and equipped with both retrospective and current examples from the Olympiad Research Project. Olympians are defined as the top-performing high-school students in mathematics, chemistry, physics, or computer science.

These authors chose students with mastery in these domains as these students have the potential to become mathematicians, scientists, and/or create patents which are critical to global economies. Sample sizes per country ranged from 6-20. Throughout the book, the authors, renowned in the fields of gifted education and cross-cultural research, refer to these samples in order to generate pedagogical and parental implications and applications. Their long-term objectives are to help cultivate technical talent on the global platform.

Particularly noteworthy is Campbell and Kirsi's recommendation for more soundly developed psychometric measures and the need for cross-cultural studies within nations in order to understand and respect differences. These differences refer to disparate mores and socialization factors among gender. The authors ably detail the derivation of their cross-cultural constructs as an integration of quantitative and qualitative (Qual/Quan) methods and provide sound examples in identifying the cultural meaning of the construct for each target ethnic group as mores influence achievement. Campbell and Kirsi adeptly point out that prior international studies, like the IEA, TIMSS, and Stevenson Studies (1983-1990), use a single item to measure a multi-dimensional psychological construct. The authors substantiate the need for multi-item psychological or socio-psychological constructs, as single-item constructs cannot adequately measure expectation, attribution, and the student's desire to excel in a specific technical domain. They also point to the need to conduct additional research with the current large-scale cross-national studies, PISA (2000), and TIMSS-R (1997), as interpretation of data within each cultural framework were omitted.

As a published author on secondary analyses of large-scale secondary databases, NELS: 88 and the TIMSS-R, respectively, I concur with the need to develop multi-item psychological or socio-psychological constructs, as a single item cannot accurately measure multi-dimensional constructs. I discovered on several occasions in both the science and mathematics student questionnaires of the TIMSS-R, only one source variable/item was used to measure important attributional variables. One prominent example was the source variable SQ2-6g (also embedded in the derived variable BSDMDAY7) in which students report on hours spent each day studying or doing mathematics. Derived variables were formed by merging and rescaling source variables within the student questionnaires (Gonzalez & Miles, 2001), Within the variable BSDMDAY7, the hours offered were intervals between 0 and 7+ hours, with one selection offering a large span of "3-5 hours;" potentially leading to inaccurate interpretation by both the student and researcher. For example, if some students maintain they study 2.5-2.75 hours on average, they might disregard the prior selection of "1-2 hours," and choose the next choice of "3-5 hours" which almost doubles their true studying time; thus evincing both incorrect data and interpretation by the researcher.

Several chapters address the complexity of ensuring that cross-cultural instruments are accurately translated and representative of the constructs on family processes and attribution, and cite conceptual models for cross-cultural application. Of particular note is Campbell and Koutsoulis's blueprint for establishing the validity of constructs. They address the salient dilemma in cross-cultural research as understanding and translating the different languages, their nuances, and colloquialisms, and provide a list of guidelines and recommendations. Heller and Perleth substantiate the critical necessity of testing the reliability and validity of the constructs through discussion of the Munich Model of Giftedness and Talent which was developed in the Munich Giftedness Study (1985-1997). Guthery and Lowe (1995) provide several techniques and examples in instrument translation. Feng persuasively discusses the efficacy of the forward and back-translation and the lack of semantic equivalence, and provides recommendations for future cross-cultural researchers.

The concluding chapters restate the validity of the Olympiad Research Project: its close association to the Terman longitudinal studies of the twentieth century involved in the adult productivity of scientists, engineers, and mathematicians; latest Bayesian statistics; and optimal learning environments. Particularly notable is Nokelainen and Tirri's pragmatic and comprehensible outline (equipped with concrete examples from the ORP) for variable selection based on Bayesian classification modeling and inspection of probabilistic dependencies between the variables with Bayesian dependence modeling.

In summary, Cross-cultural Research: Basic Issues, Dilemmas, and Strategies is a significant contribution to the field of international research and sure to ignite controversy. This book is essential for policy makers and cross-cultural researchers as it details the weaknesses of previous large-scale international studies and provides a blueprint for conducting accurate international research. Through concrete examples from the "Academic Olympians Research Project," the well-known international researchers provide strategies for validating constructs, procedures for creating valid and reliable cross-cultural instruments, solutions to common problems in international research, and innovative and robust statistical methods for analyzing cross-cultural data.

REFERENCES

Gonzalez, E. J., & Miles, J. A. (Eds.). (2001). TIMMS 1999 User Guide for the International Database. Chestnut Hill, MA: Boston College.

O'Connor, S. A., & Miranda, K. (2002). The linkages among family structure, self-concept, effort, and performance on mathematics achievement of American high school students by race. American Secondary Education, 31(1), 72-85.

O'Connor-Petruso, S. A., & Miranda, K. (2004). Gender inequities among the top scoring nations, Singapore, Republic of Korea, and Chinese Taipei, in mathematics achievement from the TIMSS-R study. Proceedings of the IRC-2004 TIMSS, 2, 31-46.

Scott, E. (2005). Comparing NAEP, TIMSS, and PISA in mathematics and science. Retrieved November 10, 2005, from the National Center for Education Statistics Web site: http://nces.ed.gov/timss/pdf/naep_timss_pisa_comp.pdf

Reviewed by Sharon Anne O'Connor-Petruso. Sharon is an Assistant Professor for Mathematics, Science, & Technology at SUNY College of Old Westbury, NY. She is actively involved in AERA's International Studies SIG and is known for her work in secondary analyses of large databases. E-mail: [e-mail]