In global selection campaigns, does the language of administration make a difference?

The world of high finance has become dominated by the competitive need to attract talent. Ten years ago, this “war for talent” was exposed as a strategic business challenge and a critical driver of corporate performance (Michaels, Handfield-Jones & Axelrod, 2001) and this war doesn’t seem to be abating. In 2008, a large proportion of “baby boomers” are set to retire, therefore, there will be fewer graduates with the right critical skills entering the market who can act as ready replacements for the skilled and experienced leavers.

Large organizations are increasingly looking worldwide for their talent (China and India are two of the major sources) but even these relative new sources of talent are being taxed. Not long ago, India’s skilled labor supply seemed limitless, but in 2006, less than a quarter of companies recruiting in India said they were meeting recruiting needs (Business Week Online, Aug 2007). By 2010, it is predicted that India will face a shortfall of 500,000 staff capable of doing work for multinationals. The ability to attract the right candidate and to assess that talent as quickly as possible has become a business necessity. The Internet is proving to be the most effective, cost-efficient way of sourcing and engaging that potential (Weiss & Barbeite, 2001).

Part of that engagement is the recognition that applicants may prefer to complete their online applications in a language other than English. Organizations are increasingly asking test providers to make their tests available in a variety of languages, in order to facilitate applicants outside English-speaking countries and provide a more equitable test of ability, not mediated by specific English language proficiency. The British Psychological Society’s guidelines for the development and use of computer-based assessment (British Psychological Society Psychological Testing Centre, 2002) stresses the importance of the recruitment process (and computer-based assessment, in particular) as a transaction between an employer and a candidate where there is a need to treat the candidate with respect. A major part of this is the candidate’s perception and experience of the assessment; image is important to employers who wish to attract the best candidates (Scheu, Ryan & Nona, 1999), especially as all candidates are also potential customers. Accordingly, a global image and attraction strategy needs an appreciation of the range of languages of potential candidates.

Language of test administration, as an issue, has tended to be addressed in the educational testing field where testing is often for diagnostic and developmental purposes. Nevertheless, there has been some investigation of testing language in high-stakes educational selection settings, especially surrounding educational selection using tests such as the SAT (Angoff & Cook, 1988; Schmitt, Dorans, Magrina & Cook, 1998; Cascallar & Dorans, 2003) or the GRE (Pennock-Roman, 1998). In the last study, which involved Puerto Rican students who were more proficient in Spanish than in English, major differences in students’ test performance were found (especially on the GRE Verbal test), which paralleled their English language proficiency.

Research has been conducted to study differences in personality scores due to language (mostly with bilingual groups). In one study, the participants took the same questionnaire in both English and Spanish; there were consistent differences dependent on the language, in which they completed the questionnaire in (Whitworth, 1988).

The equivalence of tests in different languages is quite difficult to estimate and this equivalence is greatly affected by poor or partial translation. Van de Vijver and Hambleton’s initial work (1996) on providing a framework for translating tests for use in cross-cultural settings outlines the various biases and problems associated with achieving true equivalence; this work has been more clearly set out in the International Test Commission (2006) guidelines on computerized test use.

Existing research suggests that candidates are happier taking a test in their own first language and bodies such as the American Educational Research Association, American Psychological Association, the British Psychological Society and the Joint Committee on Testing Practice (2000) suggest that this is highly desirable. It is obviously the case that candidates would be happier, but the implication is that the candidates taking tests in a language other than their own language will be less happy and possibly less attracted to the organization involved.

This, in fact, may not be the case because the choice of test language is rarely in the control of the test taker. Part of a test taker’s expectations might be that an organization based in a predominantly Englishspeaking country would expect applicants to complete tests in English. While most employers will still use a test, or series of tests, in a single language, the use of Internet testing has allowed greater choice and flexibility in the nature of tests administered and the range of languages available, and has allowed the test taker more control of the testing scenario than ever before. This situation has changed due to the globalization of recruitment campaigns and the ease with which suitable applicants can make an impression on recruiters who are potentially thousands of miles away on another continent.

For the major multinationals, large accounting and consulting organizations, and some major public service employers, the volume of applicants from outside the jurisdiction is increasing massively. The use, sophistication and range of online psychometric instruments available to help “sift out” less suitable candidates is likely to increase and the need to accommodate a greater range of candidates becomes more imperative. However, this does not mean that the overall standard of applicants is necessarily rising. Bartram (2001) talks about “kissing lots of frogs before you find your prince;” the Internet is helping employers meet more “frogs” than ever in the hope of finding that “prince.”

Kenexa is addressing this increasing demand for tests in languages other than English by developing a range of psychometric assessment tools in a variety of languages including, French (European and Canadian), Spanish (Latin American and Castilian), German, Italian, Portuguese (European), Polish, Russian, Turkish, Japanese and Chinese (Simplified Mandarin). Recent developments include the creation of a Chinese language personality questionnaire.

This list is increasing and is likely to increase further as more and more organizations are looking for assessments in a wider range of languages and where Kenexa operations expand into many more countries.

References

• American Educational Research Association, American Psychological Association, and National Council on Measurement in Education (1999). Standards for educational and psychological testing. Washington, DC: American Psychological Association.
• Angoff, W. H. & Cook L. L. (1988). Equating the score of the Prueba de Aptitude Academica and the Scholastic Aptitude Test (College Board Report No. 88-2). New York: College Entrance Examination Board.
• Bartram, D. (2001). Internet recruitment and selection: Kissing frogs to find princes. International Journal of Selection and Assessment, 8, 261-274.
• Bartram, D. & Hambleton, R. K. (Eds.) (2006). Computer-based testing and the Internet: Issues and advance. John Wiley & Sons, Chichester. Business Week Online (Aug 2007) India’s talent get loads of TLC. Retrieved Sept 10, 2007 from http:://www.businessweek.com/magazine/content/07_34/ b4047406.htm
• Cascallar, A. S. & Dorans N. J. (2003). Linking scores from tests of similar content given in different languages: Spanish language PAA and English language SAT (College Board Report No. 03-20). New York: College Entrance Examination Board.
• International Test Commission (2006). International guidelines on computer-based and internet-delivered testing. International Journal of Testing, 6, 143-171.
• Joint Committee on Testing Practices (2001). Rights and Responsibilities of Test Takers: Guidelines and Expectations. Washington. DC: Joint Committee on Testing Practices.
• Michaels, E, Handfield-Jones, H. & Axelrod B. (2001). War for Talent. Boston. MA: Harvard Business School Press.
• Pennock-Roman, M. (1998). Measuring developed academic abilities using Spanish vs English-language tests: PAEG/GRE relationships for Puerto Ricans who are more proficient in Spanish than in English. (GRE Research Report No. 98- 40). Princeton, NJ: Educational Testing Service.
• Scheu, C, Ryan, A. M. & Nona F. (1999). Company web sites as a recruiting mechanism: What influences applicant impressions? Paper presented at the 14th Annual Conference of the Society for Industrial and Organizational Psychology, Atlanta, GA.
• Schmitt, A, Dorans, N. J., Magrina, A. & Cook L. (1998). Predicting scores on the English language SAT from the Spanish language PAA and the Spanish language English as a Second Language Achievement Test. Paper presented at the annual meeting of the American Educational Research Association. San Diego, CA.
• Van de Vijver, F. & Hambleton, R. (1996), Translating tests: some practical guidelines. European Psychologist, 1, 89–99.
• Weiss, E.M., & Barbeite, G.F. (2001). Internet as a job source and job site preference. Paper presented at the 16th Annual Conference of the Society for Industrial and Organizational Psychology, San Diego, CA.
• Whitworth, R. H. (1988), Anglo- and Mexican-American performance on the MMPI administered in Spanish or English. Journal of Clinical Psychology, 44(6), 891-897.