Test and Measurement. Bias and Cultural diversity in Psychological Assessment

ABSTRACT

A century of research evidence of psychological assessment shows that scientist have made several efforts to advance a ‘‘culture free’’ tests (Jensen, 1980). Research also demonstrates that only a few numbers of issues in psychology research divide researchers and the general public as the use of standardized assessments with diverse culture. This article uses empirical evidence to analyse cultural bias in psychological testing and explores various approaches that described and examine bias in psychological assessment. To illuminate these concerns and possibilities in a concrete context, the article analyse the history of psychological assessment and explains the application of psychometric and socio-cultural framework for psychological testing. This paper explores and discuses: (1)the taxonomy of bias and equivalence in psychological testing,(2) identifies issues surrounding test bias, (3)explain sources of bias (4) evaluate how culture influences psychological assessment of diverse groups and last but not the least,(5) examine the inference of bias controversy and recommend various processes that remove bias in psychological assessment. The study conclude that psychological test performance on different cultural group shows different results.

KEYWORD: Bias, Cultural diversity, Method bias, Psychological assessment

Introduction

Psychology as a discipline cannot achieve the inevitability and precision of the physical sciences, except it leans on research and measurement. This development prompts the need to introduce a series of psychological assessments and measurements in psychological research. Research shows that possibly no other assessment problem is as intense, contentious, and often discussed as that of bias in psychological testing. Also, at the moment only a few numbers of issues in psychology research divide clinicians and the general public as the use of standardized assessments with diverse culture. The dominant subject, concerning clients, parents, and clinicians are the long-standing consequences that happen when mean test outcomes varies among ethnic group i.e. Blacks, Hispanics, American Indians, Asian Americans, to mention a few. This continues to be a major discussion in psychological research till date, particularly, how test bias and cultural diversity influence assessment.

A century of research evidence of psychological testing shows that scientist have made several efforts to advance a ‘‘culture free’’ tests (Jensen, 1980). The history of psychological assessment confirms that diverse efforts were made in the past to create ‘‘culture-free’’ methods (Anastasi, 1988; Cattell, 1940). For instance, reports show that the consequence of culture can be eliminated or measured when spoken items were removed, and non-verbal, items were used in testing. However, this notion turned out to be wide of the mark. The review of literature establishes that scientists using a widespread diversity of cultural groups across the world have occasionally detected greater group variability in performance and additional non-verbal assessments than what is recorded in oral assessments (Anastasi, 1988; Irvine & Berry, 1988). These efforts corroborate the existing fact that ethnic/cultural subgroup variation in standardized assessments (Herrnstein & Murray, 1994). Nevertheless the explanations for these differences remain a controversy till date. Based on this premise, psychological research is totally incomplete without the application of psychological testing.

Also, more than any other professionals, psychologists are the main user of psychological tests in research. This process includes: intelligent test, aptitude test, personality test, cognitive test and mental test to mention a few. On the other hand, the debate about how tests influence the judgments of test administrator and the prospects of the testee remain inconsequential. For example, “an intelligence test is an unbiased, insignificant instrument until somebody allocates meaning to the outcomes derived from it. As soon as a meaning is attached to somebody's score, the person will experience different consequences, stretching from artificial to natural life time changing. These can be reasonable or biased, supportive or injurious, suitable or ill-advised,subject to the connotation attached to the test score”.Similarly, the review of literature confirmed that neuropsychological procedures filed to identify an appropriate analytical accurateness when administered to non Caucasian, cultured natural English-speaking, as well as those from middle and upper socioeconomic status (Ardila, Rodriguez-Menendez, & Rosselli, 2002; Boone, Victor, Wen, Razani, & Pontón, 2007; Brickman, Cabo, & Manly, 2006; Loewenstein, Arguelles, Arguelles, & Linn-Fuentes, 1994; Manly, 2005). Also, research continuously make known that not only is this concept generate controversy, but that psychological testing amongst ethnically and linguistically varied people is seriously susceptible to hypothetical and experimental basis in neuropsychological practice.

The purpose of the paper

This essay analyses the empirical evidence about cultural bias in psychological testing and exploring various approaches that described and examine bias in psychological assessment. The essay begins with brief overviews of history of psychological assessment and explains the application of psychometric and socio-cultural framework forpsychological testing. Also, the paper will look at the taxonomy of bias and equivalence in psychological testing and identifies issues surrounding test bias, sources of bias and how culture influences psychological assessment of diverse groups. Lastly, the paper will examine the inference of bias controversy and recommend various processes that remove bias in psychological assessment.

Historical Perspectives

Psychological testing in its current arrangement originated a little more than a century ago in a research laboratory of sensory discernment, motor skills, and reaction time. Research shows that Francis Galton (1822–1911) designed the first series of tests, an odd variety of sensory and motor methods. James McKeen Cattell (1860–1944) an American psychologist having worked with Galton announced the recent testing outline in his model paper entitled “Mental Tests and Measurements.” in 1890.The paper indicates that the outcomes of the outline would be of huge systematic importance in determining the reliability of intellectual developments, their interdependence, and their difference under diverse situations. It also explains that the assessments will be thought-provoking, and, possibly, valuable with respect to teaching, way of life or sign of illness. In addition, the logical and practical worth of such assessments in a live setting would be substantially augmented if an identical scheme is embraced and the resolutions made at various periods and situation is likened and shared (Cattell, 1890). Based on this premise, the assumption that “possibly” the assessment would be helpful in “teaching, method of life or sign of illness” is noted as one of the far-sighted understatements of all time.

Research on cultural bias in psychological assessment has raised more argument than any other issue in psychological research. This issue continues to generate debate since the beginning of nineteen century when the first intelligence scale was developed and Stern presented measures to assess intelligence (Binet & Simon, 1916/1973; Stern, 1914). This skirmish is not restricted to intellectual aptitude tests, and it continues to lure the attention of the general public till date. Numerous scholars and researchers have come up with publications that addresses the issue, as this turns out to be contentious (Gould, 1996; Herrnstein & Murray, 1994). The controversy surrounding intelligence tests also attracts many court cases, incites government regulation, and bringsflay from general media (Brown, Reynolds, & Whitaker, 1999; Reynolds, 2000a). Jensen (1984a) maintains that the conduct of test bias prior to the 1970s was disconnected, haphazard, and theoretically muddled.

Evidence shows that the perfectly and commonly agreed-upon connotations of bias were not there. This includes a psychometrically strong procedure that empirically identifies test bias. This leads to the cautiously thought-out justification and numerical procedure that psychometric had long advanced in the areas like consistency, validity, and item selection, (p. 507). Interestingly, the commencement of the early seventy, witnesses an enormous bulk of unbiased and experimental studies of multifaceted problems such as the development and usage of standardized psychological assessments with American-born, English-speaking subclasses (e.g., see Berk, 1982). Besides, one general fact and age-old claim in psychological testing is that mean variances are as a result of “cultural bias” in assessment. Arthur Jensen (1980) systematically reviewed the experimental studies on test prejudice and he concludes that test predisposition is the most shared gathering opinion of critics. Based on the empirical measures, Jensen (1980) establishes that standardized ability tests forecast correspondingly for English-speaking majority, American-born, and minority subclasses, at the same time measures related concepts.

Also, Gregory (2004)corroborates the earlier research that no exercise in contemporary research is more attacked like psychological testing. Research shows that analysts maintain an exceptional and often passionate criticism of aptitude testing in particular to date. The argument and discussion is rooted in perceived variances of typical intelligence scores particularly, among several ethnic clusters (Blacks) and racial clusters (immigrants) in the early nineteen century (Cole & Zieky, 2001). Research shows that African Americans score, 15 points smaller than white colleagues on traditional aptitude assessments with extraordinary language/spoken and traditional loadings (Flanagan & Ortiz, 2001). Therefore, conclusion on disparity group test score performance in aptitude test intensified debate on test bias (Gregory, 2004).

Psychometric framework.

The issue of preconceived notion is basically a valid subject in analyzing psychometric concepts. A test is not reliable if it shows inaccuracies in context when applying to recognizable subpopulations. Research confirms that the tasks to test objectivity are developed between the late 60’s and early 70’s as a consequence of a number of issues coming together. Some of these issues include: evolving anticipation of fairness in research outcomes; the emerging opinion that African-American beliefs are the same as leading white values; and the ensuing authentication of methods that accentuated its characteristic potentials. Nevertheless, the most significant factor surrounding this progress is linked to the advent of black psychology. This development measures and analyzes Afro-Americans principles (Hilliard, 1995). Moreover, black psychology believes that appraisal need to be articulated through ethnic material acquainted to the assessment participant, and that a test must give consideration to histories of domination, as well as emotional persecution. Undoubtedly, such mind-set may possibly not depend on apparatuses established in the old model. Based on this resolves, most consideration for assessment is directed on psychological assessments and traditional predispositions such as intelligence assessments or personality inventories. However, this consideration without doubt spilled over into accomplishment testing.

Socio-cultural framework

A substitute opinion of the psychometric method sees assessments as a portion of the social marvels of public education. It indicates an evidence of larger social and ethnic problems. Frisby (1998) enunciates three probable expert methods to explain culture: i.e. the practitioner-clinician, socially conscious advocate and theorist-researcher. This notion proffers understanding and superior answer to questions of prejudice in testing as psychometric examination failed to place our effort in a social context. However, because of poor knowledge of the cultural context, it is difficult as researchers to solve the issue of bias in testing. Although research acknowledged the fact that psychometric studies notices the presence of bias in testing but find it hard to clarify. Thus, the theorist - researcher tries to comprehend the causes of performance through unbiased progress and appraisal of the concept. The theorist-researchers investigate the interactions between smaller group affiliation and educational achievement as steered mainly by empirical data. Based on the fact that they are probing culture, they find it hard to comprehend and therefore, observes assessment data only. The research on bias starts with the set of courses then focuses on the methods of measuring information and abilities. Test data helps describing the problems, nevertheless the objective is to cultivate and improve philosophy of what and by what means does students learn and in what manner does this influenced by cultural experience. Closely tangled is the question, “What must be done to make sure that all learners study?” To answer this question, effort should be directed towards the use of assessment tools that are consistent with both the cultural experience and the anticipated result

Research shows that practitioner-clinician, must get close to the assessment director’s role, and search for consistent knowledge to guide practice. This is logically determined by theory, and focuses on practical answers to daily difficulties. The illustration appeals to both psychometric and socio-cultural agendas and makes assessment an instrument toward pupil education rather than an end in itself. For a dependable awareness that guide training, a respectable practitioner-clinician must see to it that experimental outcomes are construed within a situation that contains the culture of the learner and the social moulds of the assessment. This is based on the beliefs that a learner needs to obtain, irrespective of their cultural inheritance.

Also, socially cognizant supporter try to defend the rights that historically-excludes groups. As similar to the theorist-researcher, the campaigner’s probe the bigger culture and the learner’s place within it, then changes to what learners need to be acquainted with and the manner of measurement. Nonetheless, the advocate shows little signs of worries about fairness or model development compares to altering significant injustices of domination and segregation. Asa result, experts incline to ignore advocates for not knowledgeable in theory, but greatly obligated to their unyielding voice.

Test Bias Controversy: Definitions and Taxonom

Historically, psychology reveals quite a few examples of far-reaching overviews about variation in aptitudes and characters of ethnic people that were founded on psychometric poor processes. However, to avert creating a far-reaching declaration which in the end disfavour to the subject, a lack of bias (i.e., equivalence) ought to be established as an alternative for assuming (Poortinga and Malpass, 1986). The review of literature establishes that assessments are frequently regarded as prejudiced against Black and other racially and linguistically varied groups. This situation is counter to students from lower socioeconomic group as it favour both middle class and white learners. According to Gregory (2004) test bias is defined as “unbiased numerical manifestations that observe the patterns of assessment scores for appropriate subpopulations” (p. 242). Gregory (2004) continues by concluding that agreement occurs on the numerical standards which shows an assessment as subjective.

Bias exists when an assessment score has connotations with a related, definable subclass of test participants and varied in connotations or effects for the remnants of test participants. Hence, bias is the degree of difference in the validity of an assessment score for a definable and an appropriate subclass of assessment participants (Cole & Moss, 1998, cited in Gregory, 2004, p. 242). However, from a social standards perspective, when a test is prejudiced, the worry is synonymous to renunciation of prospect and the improper undesirable assumption. Also, there are other standings that are germane to debates concerning testing CLD groups. Research shows that while a test is not subjective in principle, it might be biased (see Cole & Zieky, 2001). Hence, assessment objectivity is basically related to the social effects of test outcomes (Gregory, 2004, p. 249; Hunter & Schmidt, 1979).

[...]