Acculturation and evaluation of Mexican Americans with disabilities.
Several issues have been raised in the rehabilitation process with minority individuals, regarding the appropriateness of traditional psychological assessment instruments. Potential cultural bias in psychometric approaches to evaluation have been reported, but often with conflicting results. Organist (1982) outlined several key variables most significant in relation to the rehabilitation assessment process: (1) Situational/Personality variables, (2) Language differences, and, (3) Time perception. These variables may bias the evaluation process in either a positive or negative direction and thereby produce undesirable consequences. Thus, there is a need for valid assessment procedures where the influence of cultural factors is minimized, and thereby, those sources of measurement error which invalidate the assessment process eliminated. Research on psychometric approaches, particularly those related to measurement of cognitive-intellectual functioning of Mexican Americans, have been equivocal in their results. Both the Wechsler scales and the Peabody Picture Vocabulary test have been studied, with reports of good reliability, but questionable validity (Budoff, Corman & Gimon, 1976; Goldman & Hartig, 1975; Henderson & Rankin, 1973; Henggeler & Tavormina, 1979; Reiber & Womack, 1968; Simon & Joiner, 1976) Criticisms of intelligence test have frequently included: (1) Items requesting information inappropriate to the individual's ethnic background; (2) Items requiring high verbal functioning; and (3) Items tapping skills not generally taught in a particular culture (Armstrong, 1972).
These previous researchers have reported their results on Mexican American samples using Spanish language use or Spanish surname as their primary cultural descriptors. While useful, these descriptors do not include other cultural factors of potential significance in measurement. Current theorists have clearly identified a number of additional acculturation variables and have developed procedures to assess them (Cuellar, Harris & Jasso, 1980; Padilla, 1980; Szapocznik & Kurtines, 1980).
Cuellar and Arnold (1988) and Rivera (1983) indicated the importance of evaluating level of acculturation in an attempt to provide a basic understanding of the client's cultural background, and thus reducing potential biases in psychometric evaluation through formal, clinically moderated approach to interpreting data.
The present study attempted to identify the relationship between certain cultural factors and assessment of rehabilitation potential in a Mexican American sample of individuals with disability.
Prior to data collection, each potential participant was approached by their Vocational Rehabilitation counselor, on the basis of Spanish surname or mother's maiden name, and their participation requested. Additional criteria for participation in the study included a primary disability designated as physical. Finally, participants who indicated they were monolingual in English or Spanish on the Acculturation Rating Scale for Mexican Americans were also excluded in order to create a sample homogeneous with regard to bilingual English-Spanish abilities. Seven participants were excluded on this basis, with six monolingual Spanish-speaking, and one monolingual English-speaking participant.
If agreeable, each participant was referred to the study and told that the research was aimed at increasing the understanding of the interaction of culture, and the vocational rehabilitation process. All subjects who were approached were in the process of being accepted for vocational rehabilitation services. Each subject then signed a bilingual agreement to participate and a release of confidential information.
A total of 37 participants from the State Vocational Rehabilitation Commission participated in the study. In this sample, 28 were male subjects, with 9 females. Data were collected from subjects along the rural South Texas borderlands areas (n=31) and also from two urban areas (n=6). Average acculturation rating was 3.03, with a standard deviation of .47, suggesting a bilingual, bicultural sample. Average age was 28.57 (S.D.=9.0). Educational level showed an average of 10.46 (S.D.=2.54). Disability groups represented included motor (n=23), neurological (n=11), and chronic disease (n=3) impairments. Ninety two percent of the sample had been designated as severely disabled by their VR counselor, utilizing state established criteria. A three month follow-up of these subjects showed no indication of a subsequent DSM III diagnosis of affective, personality or other significant psychological difficulties.
Independent variables for the present research were age, education, socioeconomic status, acculturation, and the four factor components of acculturation: (1) language preference, (2) ethnic identity and generation, (3) ethnicity of friends and associates, and (4) reading and written language and direct contact with Mexico. Background data was collected through a bilingual data sheet developed by the authors.
The major independent variable, level of acculturation, was documented by the Acculturation Scale for Mexican Americans (Cuellar et al., 1980). This bilingual English-Spanish behavior rating scale measures acculturation on the basis of the subject responses to 20 questions relating to content areas such as ethnic identification, language preference and use, peer relationships, food preferences, and generation. An average rating for each subject is obtained placing the individual on a unidimensional continuum from 1 (monolingual Spanish, monocultural Mexican) to 3 (bilingual English-Spanish, bicultural Mexican American) to 5 (monolingual English, monocultural Anglo). Developed and validated by Cuellar et al. (1980), it has since been cross-validated in an independent study by Montgomery and Orozco (1984). The results of the cross-validation study closely corresponded to the original normative data provided by Cuellar, et. al. (1980). Internal consistency of the scale was high and factor analysis yielded the same four factors identified by Cuellar, et. al.
The abbreviated battery of the McCarron-Dial Evaluation System (MDES) as described by McCarron and Dial (1976) and cross validated by Packard, Henke, and McCollum (1976) was utilized to measure the subject's level of vocational potential (McCarron & Dial, 1976). The dependent variables for the present study were the specific McCarron-Dial predictor measures. The MDES consists of four predictor measures that are grouped into three factors: (1) verbal-cognitive; (2) sensory; and (3) motor. Each of these factors are important in the assessment of behaviors that determine work potential or effective performance in a work environment. Additionally, the study included measures of adaptive behavior, and information processing. These latter measures have been recently developed for use in vocational assessment, as part of the total McCarron-Dial Evaluation System. The specific factors and measures used in predicting vocational potential were:
1. Verbal cognitive/information processing: Peabody Picture Vocabulary Test-Revised (L. Dunn & L. M. Dunn, 1981), and the Perceptual Memory Task (McCarron, 1984). The Peabody Picture Vocabulary Test (PPVT) is designed to measure an individual's verbal intelligence through the client's ability to associate a vocalized word, usually a noun or verb with a pictorial representation of the word. Each subject's verbal IQ or standard score was integrated into the McCarron-Dial Evaluation System (MDES). The Perceptual Memory Task (PMT) was developed as a standardized method of assessing information processing skills and individual learning style. Since vocational evaluation attempts to systematically describe an individual's functional capacities, the PMT can be an integral part of vocational assessment. The PMT involves measures of spatial relations, visual recognition and visual sequential memory, long term recall and verbally mediated visual memory skills. The PMT's plan was to design a task involving cognitive problem solving strategies. The PMT can be utilized with a wide range of individuals with various cultural backgrounds since instructions to the tasks are simple do not require a verbal response, and require only receptive comprehension of colors which are orally named by the examiner. The total PMT score consists of the sum of all four subtest areas, converted to one standard score. Additionally, performance on the PMT has been shown to be significantly correlated with intellectual functioning on the Wechsler scale and PPVT
2. Sensory functioning: Haptic Visual Discrimination Test, (McCarron & Dial, 1975), and the Bender Visual Motor Gestalt Test (Bender, 1938). The Haptic-Visual Discrimination Test (HVDT) is used as an assessment of the neurological integration of tactual and visual discrimination processes. The HVDT is organized to provide discrimination measures in four areas: size, shape, texture, and configuration. The HVDT requires subjects to manipulate geometrically shaped objects which are hidden from the visual field. The subject then attempts to identify the correct representation of the object on a chart. Each subject's number of correct responses in each of the four areas is incorporated into the MDES.
The Bender Visual Motor Gestalt Test developed by Lauretta Bender (1938) was designed as a test of visual-motor integration. The test consists of nine cards with abstract designs. Clients are asked to reproduce each design one at a time on a blank sheet of paper. Reproduction of the designs are influenced by integration of perceptual and motor processes. The Koppitz error system is used to score each design (Koppitz, 1975). The error score is converted to a standard score.
3. Motor functioning: McCarron Assessment of Neuromuscular Development (McCarron, 1975). The McCarron Assessment of Neuromuscular Development (MAND) provides a measure of a subject's fine, gross, and total motor abilities. Assessment of fine motor abilities is determined by five tasks, (1) beads in box, (2) beads on rod, (3) finger tapping, (4) nut and bolt, (5) rod slide. Assessment of gross motor abilities are measured by (1) hand strength, (2) finger-nose-finger, (3) jumping, (4) heel-toe tandem walk, and (5) standing on one foot. The MAND can also be used to determine bimanual dexterity, persistent control, muscle power and balance-kinesthetic control. Three standard scores are derived from the MAND corresponding to the fine, gross and total motor abilities.
4. Integration-Coping (adaptive behavior): Street Survival Skills Questionnaire (Linkenholder & McCarron, 1980).
In order to get a better picture of functional strengths and weaknesses of Mexican Americans being evaluated, it has been suggested that both measures of intellectual functioning and adaptive functioning be conducted (Mercer, 1976). A split half version of the Street Survival Skills Questionnaire (SSSQ) was utilized since the instrument has relatively high internal consistency. The shortened version thereby provided greater efficiency for data collection purposes. The SSSQ was constructed as a comprehensive measure of community relevant adaptive skills. It provides an objective measure of specific aspects of adaptive behavior.
The SSSQ consists of nine subtests, each measuring nine content areas most relevant to adaptive behavior. The nine areas are: (1) Basic Concepts, (2) Functional Signs, (3) Tool Identification, (4) Domestic Management, (5) Health, First Aid and Safety, (6) Public Services, (7) Time, (8) Money, and (9) Measurement. The sum of all nine subtest scores are converted to one standard score.
Each subject signed a bilingual agreement to participate along with a release of confidential information. Each subject was scheduled for an individual appointment at which a bilingual research assistant then administered (a) the Acculturation Rating Scale for Mexican Americans (Cuellar et al., 1980), (b) a background data sheet developed by the authors, and, (c) the abbreviated McCarron-Dial Vocational Evaluation Battery (McCarron & Dial, 1976). Each instrument was administered in English. Time of administration averaged 3.5 to 4 hours per subject.
All dependent variable measures were converted to standard scores. Multiple correlation analysis was used to examine the relationship between overall level of acculturation, the four acculturation factors, and each of the abbreviated McCarron-Dial factors contributing to prediction of vocational potential (Table 1).
Neither the PPVT nor the PMT showed a significant relationship to overall level of acculturation in this homogeneous bilingual, heterogeneous bicultural sample. However, in looking at the two measures of cognitive information processing, a stronger relationship with level of acculturation was noted for the Peabody than for the PMT. The strength of this relationship is accounted for in part, when one focuses on the specific cultural factor of ethnic identity and generation removed from Mexico. The relationship between this factor and performance on the PPVT was significant at the .05 level. The specific items contributing to a rating on this factor include: (1) "How do you identify yourself?", (2) "Which ethnic identification does (did) your mother use?", (3) "Which ethnic identification does (did) your father use?", (4) "Where were you (your parents, grandparents) born?", (5) "How much ethnic pride do you have?", and (6) "How would you rate yourself (along a Mexicanized-Anglicized dimension)?".
In the area of sensory functioning the Bender provided a correlation coefficient with acculturation approaching zero. Additionally, in the area of visual tactile functioning, the Haptic Visual Discrimination Test also showed no significant relationship with overall level of acculturation.
In the area of motor functioning, the McCarron Assessment of Neuromuscular Development had been administered resulting in summary scores for fine motor functioning, gross motor functioning, and overall total motor functioning. No significant relationships were found between motor functioning and level of acculturation in any of these areas.
In the area of adaptive behavior functioning, the Street Survival Questionnaire Survival Quotient (SQ) showed a significant correlation with overall level of acculturation. However, as with other components of this modified evaluation, each factor contributing to level of acculturation was also analyzed to identify the possibility of more subtle influences. On the SSSQ, factor 3 of the Acculturation Rating Scale for Mexican Americans, the factor measuring ethnicity of friends and associates correlated with the overall Survival Quotient. In order to understand which area of adaptive functioning might be contributing to this significant relationship, a correlation matrix with the nine tests of the Street Survival Questionnaire was constructed relating them to the overall level of acculturation and its four factor components (Table 2).
When this matrix was constructed, the subtest measuring knowledge of public services was significantly related to factor 4 (generation), and overall level of acculturation at the .05 level. Additionally, the test measuring knowledge of health and safety correlated significantly to factor 2 (ethnic identity). Neither of these two SSSQ tests was significantly correlated with education or socioeconomic status in a subsequent multiple correlation analysis.
Results of this study suggest that the modified McCarron-Dial System represents an approach to evaluation of bilingual, bicultural Mexican-Americans which does not appear to be significantly affected by cultural factors in the areas of sensory and motor functioning. By implication, this also suggests, but does not validate, that the predictive nature of the measures would be essentially the same for bilingual Mexican Americans as it is for monolingual English-speaking Mexican Americans and Anglo Americans. Additional research is needed to determine if this inference is valid. Results also suggest no cultural influence on SSSQ adaptive behavior subscales measuring Basic Concepts, Functional Signs, Tools, Domestics, Time, Monetary, and Measurements.
Results also suggest that significant cultural bias may exist on the SSSQ in the areas of Health and Safety, and Public Services, producing a low score. A low score may reflect less knowledge than is actually the case. Adaptive behavior in the area of health and safety may be culturally different and not adequately measured by this subtest. Lower acculturation level may in fact be associated with less knowledge of "majority" culture adaptive behavior in these two areas. Studies have shown low levels of acculturation to be associated with the use of alternative, nontraditional, folk-related health care practices (University of Texas, 1979, chap.3). Another more recent study relates to the underutilization of health care, and possibly other public service agencies by less acculturated Mexican Americans. Underutilization may be due to a low knowledge of public services (Wells, Hough, Golding, Burnam & Karno, 1987).
In the area of information processing, the PMT appeared to be less correlated with overall level of acculturation, and clinically, therefore, may be a better indicator of cognitive information processing abilities, as acculturation level decreases, than the PPVT. More subtle influences than language dominance may be interacting to provide less than accurate data on cognitive-intellectual functioning of bicultural clients (Hickey, 1972). Inferentially, this also suggests that as the client population moves to a more dominant Spanish language preference, a test such as the PMT which de-emphasizes language and its nuances will maintain its relevance as a measure of information processing. Whereas, a language and culturally dominant instrument such as the PPVT-R, no longer validly measures cognitive functioning, but rather reflects acculturation level or English language comprehension, at best.
Systematic research is needed to identify and clarify the relationship between cultural factors and client performance on psychometric approaches to evaluation used with Mexican Americans. This research is needed on all existing instruments, and particularly those which have been translated for use with monolingual Spanish-speaking clients.
In conclusion, results of the present study suggest that evaluators need to assess the acculturation level of Mexican American clients. Mexican Americans rated in the bicultural to monocultural (Mexican) range on the acculturation scale should perhaps be administered secondary measures of cognitive-intellectual functioning (in addition to traditional IQ measures). These secondary measures should include performance requirements de-emphasizing language and sociocultural knowledge derived as a product of majority culture assimilation. Where results between traditional and secondary measures of information processing show discrepant results, an underestimation of cognitive abilities can be prevented through a moderated clinical interpretation of results. Increased reliance on observational and historical data can also help increase the validity of interpretations in both cognitive processing and adaptive behavior areas.
TABLE : Correlations Between Independent Variables and McCarron-Dial System Instruments
TABLE : Correlationships Between Acculturation & its Four Factor Components & the Nine Tests of the SSSQ