Journal of School Psychology A u t u m n 1966 • Vol. V, ]No. I
THE VINELAND SOCIAL MATURITY SCALE: RECOMMENDATIONS FOR ADMINISTRATION, SCORING AND ANALYSIS Duilio T. Pedrini and Lura N. Pedrini The Vineland Social Maturity Scale ( V S M S ) attempts to measure social competence. The technique has a history of indebtedness (Doll, 1953), but nonethele~ it made a unique contribution in the year it was first published (Doll, 1935a, 1935b, 1935c). Personal-social maturation is an area of critical importance when one works with children and/or retardates. The American Association on Mental Deficiency in its publication A Manual on Terminology and Classification in Mental Retardation (Heber, 1961) specifically discusses the importance of adaptive behavior in an evaluation of retardation and states that "The Vineland Social Maturity Scale is perhaps the best single measure of Adaptive Behavior currently available." The purpose of this paper is to offer some critical reactions to the VSMS and to suggest some useful procedures for administration, scoring, and analysis. Administration and Scoring The V S M S may be administered according to standard procedures or according to special procedures (Doll, 1947, 1953). If administered according to special procedures, Doll says, " . . . such results should be cautiously interpreted since normative data and correlative evidence on reliability, validity, and probable error of measurement have not yet been systematically estabfished." Departures from standard procedures, then, should be avoided if possible. The standard procedure mentioned by Doll, in his standardization of the scale did not appear to include the presence of the subject under discussion. Therefore, Doll not only allows the examination of a subject in absentia (also called indirect examination), but he recommends it. In fact, he states that " . . . to know or to see the subject . . . usually prejudices the examination" (Doll, 1953). This is possible, but hopefully the skilled examiner should have had better training, and if so, could turn his knowledge to advantage. Also, the informant is a major variable of true a n d / o r error variance. H o w is one The author, Duilio T. Pedrini, received his AB and MS degrees from the Universlty of Miami and his PhD from the University of Texas. He is an associate professor of psychology at the University of Omaha, Nebraska, and director of the psychology department at Glenwaod State Hospital-School, Iowa. The co-author, Lura N. Pedrini, received her BA from East Texas State Teachers College and MA and PhD from the University of Texas. She is a lecturer at Duchesne College and Creightan University specializing in English and American literature.
14
to differentiate by ascertaining the knowledgeability (validity?) and veracity (reliability?) of the informant? How many times have we heard, "Oh, I didn't know he could do that." Also, informants are usually emotionally involved with the subjects and many lose their objectivity. We should not expect otherwise. Internal consistency procedures (within the Scale) are helpful, but external consistency procedures (seeing the subject perform) are better. The informant and/or the inteiMewer can work with the subject and thereby improve the data collection. Mter all, we are interested in what the subject does, and not just what the informant says he does. The "burden of proof" is on the interviewer. The VSMS scoring instructions are somewhat ambiguous for the "plus no opportunity" ( + N O ) value. If a client has the following consecutive scores ( + , + N O , + ) , the + N O equals one in value; if (--, + N O , - - ) , the + N O equals zero; if ( + , + N O , - - ) , the + N O equals .5. But what about ( + , + N O , +?) or (--, + N O , +?) or ( + , + N O , + ) ? Or reversals of the examples just cited? What about multiple consecutive +NO's? Doll (1964) recommends considering the range of scores prior to assigning values to + N O . In the protocols of most clients, the problems cited would rarely appear. However, with retardates, the + N O scores seem to appear more than rarely (Doll, 1953). This might be true with special-problem clients (parents or children). Doll (1953) suggests at least two plus scores for a basal and two minus scores for a maximal in each category. We would recommend at least three of each, especially when wor'ldng with retardates (or special-problem clients) for, as Doll has indicated, they appear to scatter nearly twice as nmch as his normative group. The listing of social ages in terms of years-and-months (base 12) would have been helpful, rather than in terms of years-and-decimals (base 10). Months or years-and-months are more comparable to other scales. The condensed manual of directions should include parts of Chapters 1, 4, 6, and 7 of Doll's (1953) book. At a minimum, the more detailed scoring procedures in Chapter 7 should be included in a new condensed manual. This would be very helpful to examiners who use the condensed manual during interview procedures and the book as a resource.
Item and Category Analyses Item inclusion on the VSMS ranges from 34 items at year O-I to 3 items at year IX-X to 12 items at year X X V + . Doll (1953) claims both age and point-scale advantages for the VSMS. If one evaluates 1S
the VSMS as an age scale, a better balance is needed year-by-year in terms of number and kinds of items. The VSMS scoring~system gives about one-month of social-age credit for each item at the lower levels, but about one-year of social-age credit at the upper levels. In other words, there is not enough sampling of ability at the upper levels. Hence " . . . at the pre-school level, the VSMS is fairly adequate as a measure of Adaptive Behavior" (Heber, 1961). It may need supplementing, but not supplanting. But achievement tests are definitely needed and recommended at the school-age level, and at the adult level, social and vocational judgments in a family and community context are recommended. In essence, then, the VSMS appears most useful for the pre-school child or the retardate of preschool ability. The inclusion of a separate Self-Help category is of doubtful value because the items could be included in the other categories. The VSMS categories do not necessarily measure what they state. For example, about two-thirds of the Self-Direction items deal with money. We hope that there is much more to self-sufficiency or self direction, even in a materialistic society. There are also item headings that do not specifically inform as to the content included--an aspect acknowledged by Doll (1953). Item analysis and validation through internal-consistency procedures for the normative group are relatively well done through item 89. Too much subjectivity and extrapolation take place with items 89-117. Therefore, above chronological age 15-0, the validity of the Scale for "normals" (non-institutionafized?) comes into question. In considering item analysis and validation through internalconsistency procedures for his retarded group, Doll (1953) works through item 75. For retardates, then, validity above item 75 (social age 9-4) becomes tenuous. These items (75 through 117), however, are high when one considers the retardate's range of abilities. In other words, item analysis above item 75 is not as important for retardates as item analysis for 75 and below. If the "retardate" scored much higher, would he be retarded? The questionable aspect of this kind of item analysis is the confounding of criteria and test, input and output, background and foreground. Doll uses social-age means for the retarded in contrasting item difficulty and in assessing item discrimination. The relationship is directly dependent and correlated, yet this is not taken into account statistically. In defense, one might say that life age is also related to social age and to the Scale items, so what difference does it make if one uses social-age means for the retardates? The difference is that the items directly accumulate to give social age, not life age. External criteria are better than internal criteria. 16
In considering item analysis and validation through between-group procedures, Doll (1953) contrasts items for fife-age means of "normals" and social-age means of retardates. The criticism mentioned for the item analysis of protocols of retardates applies here. Total-Scale Standardization
There is the problem of cultural bias for the VSMS (Doll, 1953). The person of low socio-cconomic status or from a culturally-disadvantaged family may not do as well on the Scale. Conversely, a person of high socio-economic status or from a culturally-advantaged family may do better. The standardization of the current form of the VSMS dates back to the middle 1930's. Doll (1936a) himself called his original standardization a "preliminary standardization" and indicates the weakness of some of the items. The original standardization group included 620 persons from the environs of Vineland, New Jersey only. Many studies on the VSMS have been published and are listed in Buros (1953) and in Doll (1953), but a re-standardization has not been attempted. In the event of a re-standardization, deviation social-quotients rather than ratio social-quotients should be computed. For the 620 "normals" in the original standardization group, the mean socialquotient ranged from 80-112 points, and the standard-deviation socialquotients ranged from 6 to 50 points (Doll, 1953). Even if one ruled out infants below one year of age, the mean social-quotients ranged from 95 to 112 points, and the standard-deviation social-quotients ranged from 6 to 17 points. Social-quotients between ages need to be relatively equated in value, both for means and standard deviation if the results are to be comparable. The VSMS and Other Scales The VSMS, despite its limitations, is an excellent clinical technique. It is more clinical than psychometric in nature, even though results are expressed quantitatively. It is more than a questionnaire and more than a rating scale. It can serve as an interview and behaviorobservation scale. The interview technique is ideal for obtaining data relevant for counseling and remediation, and the behavior-observation technique is ideal for assessing interview-validity and social-interaction. The VSMS has inspired many other scales. Mecham has extended the Communication category of the VSMS into the Verbal Language Development Scale (Buros, 1961 ) for an evaluation of language from birth to 15 years of age. Banham developed a Maturity Level [or School Entrance and Reading Readiness scale for kindergarten and first grade, 17
regardless of high or low chronological age (Buros, 1953) as well as the Social Competency Inventory [or Adults, a behavior checklist which appears suited for the retarded or senile person (Buros, 1961, 1965). TABLE 1 Suggested Order of Discussion For Vineland Areas
1. The psychologist talks with the informant and works with the client in an informal setting. A verbatim account of the interaction should be recorded. As much as possible, the examiner should be less directive rather than more directive. 2. In lieu of method one, the Vineland should be administered as a point scale with the areas in the following order: (a) For a child, adolescent, or adult: I, II, III, IV, V, VI, VII, VIII. (b) For an infant: III, IV, VI, V, L, II, VII, VIII. The discussion should begin in general terms. If an examiner has to be specific too often, the collection of data is forced and mechanical. Rather than conversational, the data collection becomes question-and-answer, and this does not allow for spontaneity. A verbatim account of the interaction should be recorded and this includes the examiner's statements, not just the informant's answers. A "short-hand" system of recording would be very helpful. Area VIII does not exist p e r se; it is a reminder to ask "Are there, any other things that you wish to mention about "; Does he present any (other) special problems?" I. Drinking, Eating (twelve items): 11, 25, 39, 16, 20, 28, 30, 33, 38, 62, 67, 75 II. Dressing, Cleansing (fifteen tiems): 21, 37, 35, 51, 50, 52, 40, 42, 47, 54, 70, 86, 64, 74, 65 III. Moving, Walking (sixteen items): 2, 5, 3, 6, 13, 8, 12, 9, 15, 18, 23, 26, 29, 32, 45, 41 IV. Communicating; U, S, R, W;a (sixteen items): 1, 10, 17, 31, 34, 44, 58, 63, 66, 79, 73, 78, 81, 84, 90, 91 V. Playing, Working (twenty-two items): 7, 36, 19, 55, 22, 43, 57, 71, 82, 24, 48, 72, 80, 89, 98, 106, 108, 111, 113, 114, 116, 107 VI. Relating, Socializing (seventeen items): 4, 14, 27, 46, 49, 56, 59, 68, 69, 85, 88, 103, 104, 109, 110, 115, 117 VII. Self-Directing, Buying (nineteen items): 53, 61, 77, 83, 92, 93, 96, 99, 60, 76, 87, 94, 95, 100, 102, 97, 101, 105, 112 aThe denotations U, S, R, W stand for understanding, speaking, reading, =nd writing.
18
The Cain-Levine Social Competency Scale (Cain. Le~dne, and Elzey, 1964) is designed for use with the trainable retarded of chronological age five through 13. At this time, none of the above scales compare to the VSMS in standardization, validity, and reliability. At best, they can be regarded as exploratory scales, yet to be proved useful in clinical work and research.
Profile Analys;s The notion of profile analysis of the VSMS is not new as a general concept for psychometric and projective testing nor as a specific concept for the VSMS. In his book, Doll (1953) includes a profile that was developed by Myer; another is presented by Iscoe (1960). Table 1 ~ves an overview of a recommended interview procedure among and within areas of the VSMS. The word "areas" is used since the Self-Help category has been eliminated and items shifted. Doll (1953) ,gives a few examples of interview-discu~ion, and his format should be followed in questioning. The VSMS interviewer would do well to formulate general-discussion questions dealing with specific series of items. This technique is easier to learn by example, rather than by precept. An approximate level for discussion may be quickly established in an area depending upon apparent abilities. Then, the interviewer generally proceeds up and/or down to establish a basal and a maximal level within the area (or category). We generally recommend, as mentioned previously, basals and maximals of three items rather than two. Following general psychometric procedure, starting around an assumedbasal is preferable to starting around an assumed-maximal. One problem in the item seriation listed in Table 1 is the occasional inclusion of a high-level item among low-level items or vice-versa. The single item in question might establish a response-set that would not be appropriate. This is easy to overcome, however, by establishing referents as to item-difficulty--often done by the informant. Only ff the informant appears to misinterpret what is intended does the interviewer have to re-establish the appropriate referents. The knowledgeability and veracity of the informant should be evaluated not just for the overall Scale but for areas. One should check enough items in areas to ascertain objectivity.
References Buros, O. K. (Ed.) The fourth mental measurements yearbook. Highland Park, N. J.: Gryphon, 1953, Pp. 161-163. (Also see other yearbooks: 1938, 1949, 1959, 1965) Buros, O. K. (Ed.) Tests in print. Highland Park, N. J.: Gryphon, 1961. 19
Cain, L. F., Levine, S., & Elzey, F. Cain-Levine Social Competency Scale. Palo Alto: Calif.: Consulting Psychol. Press, 1964. Doll, E. A. A genetic scale of social maturity. Amer. J. Orthopsychiat., 1935, 5, 180188. (a) Doll, E. A. The measurment of social competence. Proc. Amer. Ass. merit. De[it.: 1935, 40, 103-123. (b) Doll, E. A. The Vineland Social Maturity Scale. Train. sch. Bull., 1935, 32, 1-7, 2532, 48-55, 68-74. (c) Doll, E. A. Preliminary standardization of the Vineland Social Maturity Scale. Amer. ]. Orthopsychiat., 1936, 6, 283-293. Doll, E. A. Vineland Social Maturity Scale: manual o[ directions. Minneapolis: Educat. Test Bureau, Amer. Guid. Service, 1947. (Also see 1965 edition) Doll, E. A. The measurement o[ social competence: a manual ]or the Vineland Social Maturity Scale. Minneapolis: Edueat. Test Bureau, Amer. Guid. Service 1953. Doll, E. A. Personal communication, 1964. Heber, R. (Ed.) A manual on terminology and classification in mental retardatior~. (2nd ed.) Amer. J. mont. De[ic., 1961. Iscoe, I. A profile for the Vineland Scale and some clinical applications. ]. clin. Psychol., 1960, 16, 14-16. O T H E R PUBLICATIONS The following publications have been received at the Editorial Office of the Journal. Inquiries and orders should be addressed directly to the publisher.
Behavior Problems Aichhorn, A. Delinquency and child guidance: selected papers. New York: International Universities Press, 1964. Conger, J. J. & Miller, W. C. Personality, social crass and delinquency. New York: Wiley, 1966. Milton, O. Behavior disorders: perspectives and trends. New York: J. B. Lippincott: 1965. Quay, H. C. Juvenile delinquency. New York: Van Nostrand, 1965. Smith, Judith M. & Smith, D. Child management: a program [or parents. Ann Arbor, Mich.: Ann Arbor Publishers, 1966.
Exceptional Children uGifted, Creative
Anderson, H. H. Creativity in childhood and adolescence Palo Alto, Calif.: Science and Behavior Books, 1965. Goldberg, Miriam L. Research on the talented. Columbia University: Bureau of Publications, 1965. Hildreth, Gertrude I-I. Introduction to the gi[ted. New York: McGraw-Hill, 1966. Taylor, C. W. & Williams, F. E. Instructional medka and creativity. New York: Wiley, 1966.
mMentaUy Retarded, Braln-lnjured Berlin, I.N. & Szurek, S.A. (eds.) Learning and its disorders. Palo Alto, Calif.: Science and Behavior Books, 1965. Carter, C. H. Handbook o[ mental retardation syndromes. Springfield, Ill.: Charles C. Thomas, 1966. Cruickshank, W. M. (ed.) The teacher o[ brain-injured children: a discussion ol bases [or competency. Syracuse, N.Y.: Syracuse University Press, 1966. Ellis, N. R. (Ed.) International review o[ research on mental retardation, Vol. 1. New York: Academic Press, 1966. Ifellmuth, J. (ed.) Learning disorders, Vol. I ~: l I (Special Child Publications). Seattle, Wash.: Seattle Scguin School, 1965 & 1966. Peter, L. J. Prescriptive t~aching. New York: McGraw Hill, 1965. Schuhnan, J. L., Kaspar, J. C. & Throne, Frances M. Brain damage and behavior: a clinical-experimental study. Springfield, I11.: Charles C. Thomas, 1965.
mOeaf Fnrth. II. G. Thinking without language; psychoIogiaal implications o] deafness. New York: Free Press. 1966.
20
