Yanon Volcani, Ph.D.

Psychological Testing Services International
1224 Prospect St., Ste. 110
La Jolla, CA 92037
www.psytesting.com

ABSTRACT

This is a description of the development of the first apperception test administered over the Internet – the Self Administered Global Apperception Scales™, or SAGAS™ . A brief history of apperception tests will be provided, followed by a discussion of the formation of the current test, the computerized scoring system utilized, and the process used to obtain normative data. Uses for SAGAS™ will be delineated, along with future directions.

Background

Apperception tests became a staple of psychometric assessment soon after the Thematic Apperception Test (TAT) was introduced by Henry Murray in the 1930s (Morgan & Murray, 1935; Murray, 1938 & 1943). Indeed, along with the Rorschach and MMPI (now MMPI-2), the TAT is one of most com-monly utilized assessment instrument for clinical, forensic and research pur-poses (Rossini & Moretti, 1997). While the TAT has been seen by some as more suspect than “objective” tests in terms of the rigors of its psychometric credentials, there is a long history of empirical evidence as to its validity on a variety of motives and skills (see summaries by Smith, 1992; Winter, 1996; 1999).

Motives include Achievement need (“n Achievement”), which has been found to predict entrepreneurial behavior and economic status (McClelland, Atkinson, Clark & Lowell, 1953; McClelland & Winter, 1969); n Power, which has been related to leadership skills and career position (Veroff, 1957; Veroff & Veroff, 1972); n Intimacy, which has been associated with general life adjust-ment (McAdams & Vailiant, 1982); and problem-solving abilities, found to be related to positive self- concept and inversely related to psychological distress (Ronan, Colavito & Hammontree, 1993).

Apperceptive tests also have a long history in the assessment of children.

The best known is probably still the Children’s Apperception Test (Bellak, 1954; and Bellak & Bellak, 1949 and 1975). However, rigorous validity/reli-ability properties have been better established for the Roberts Apperception Test for Children (McArthur & Roberts, 1982), the Children’s Apperceptive Story-Telling Test (Schneider, 1989), and the Tell-Me-A-Story test (Costantino, et.al., 1988). Not only do these measures assess both clinical and adaptive functioning in children, but the latter test in particular emphasizes picture stimuli and dimensional norms for minority children.

Hence, within a variety of areas of inquiry, analysis of apperceptive material has been empirically found to validly assess a broad range of affective, motivational, and cognitive capacities in adults and children.

DEVELOPMENT OF THE SAGAS™

Rationale One of the continuing limitations of the TAT is that – despite various efforts (e.g., Eron, 1950; Holt, 1978; and Murstein, 1972) – sufficient normative data across age, gender and ethnic groupings has not yet been accumulated. In addition, there has never been established a standardized procedure or specific set of cards to be utilized. Nor has there been an agreed upon scoring system, despite those previously noted for specific areas, and more global systems originally developed (e.g., Aron, 1949; Murray, 1943; Murray & Bellak, 1941; Tomkins, 1947). Indeed, in a chapter on a recent book on the TAT, Abrams calls for “…the use of a standard administration, standard card sequence, and standard scoring approach with increasing objective scoring research studies; establishment of age, gender, socioeconomic, culture, and other norms; and reliability and validity studies…” in the future use of the TAT (Abrams, 1999, p. 153). Lastly, the cards, themselves, have been criticized for being anachron-istic and overly gloomy limiting the range of affects, motives, and capacities able to be assessed.

Given the unique facets of the personality illuminated by apperceptive material, and the particular importance of gaining a quick understanding of the client’s relevant psychological themes in this HMO-driven world of brief therapy (e.g., Bellak, et.al., 1992), developing an apperceptive test fitting the delineated criteria would be helpful to clinicians and behavioral scientists.

Furthermore, it occurred to the author that apperceptive tests were particularly well-suited to be administered over the Internet. Indeed, such an environment would allow for a standardized set of stimuli to be presented in an identical fashion with standardized instructions. In addition, the presentation would be free of any of the potential biases introduced by the instructional set or gender, race and/or other specific characteristics of the test administrator. These latter factors have been shown to influence responses (e.g., Gottschalk, 1995). There would also seem to be something in the anonymity of the setting – the individual facing the computer screen– that might be conducive to candor in responding. In this sense the client/responder would feel free from any presumed judgments by the test administrator. Lastly, the Internet’s freedom from geographical boundaries, and the vast population accessibility, would allow for unprecedented possibilities in obtain-ing sufficient normative data and doing ongoing research in the area of apper-ception.

Selection of Pictures Selection of fifteen photographs for the current test focused on stimuli that would be relevant across gender, race, and age, and that would elicit mean-ingful and divergent psychological themes. Over 65,000 clip-art photographs and photographs in the public domain from photographic journals were review-ed. The fifteen selected photographs – black and white both for mood and easy downloading - were balanced as to gender; showed figures of ages ranging from children to elderly; showed figures alone and in interaction with others; and depicted figures from each major ethnic grouping (African-American, Asian, Caucasian, and Hispanic).

A trial online story sampling utilizing the selected pictures was accom-plished through the use of colleagues and friends. The resultant stories to each of the photographs produced themes that were both similar enough to form an expected response set for each photograph, and yet diverse enough as to reflect individual differences along a variety of measures.

Development of Norms

Once the trial administration was completed, notices requesting participation were placed in college and local papers throughout the major areas of the country – the East (e.g., New York, Massachusetts), the South (e.g., Florida, Mississippi), the Midwest (e.g., Michigan, Iowa), the Southwest (e.g., New Mexico), and the West (e.g., Washington and California). The announce-ment requested participation in the development of a new means of assessing emotional and cognitive functioning by way of the Internet. Participants would receive a free “psychological profile” once norms were established on the various scales measured.

In addition to producing stories (as directed by question prompts below each photograph) to the SAGAS™ photographs (www.psytesting.com), each subject completed a self-descriptive inventory (e.g., listing three adjectives that described themselves; listing their three best and worst traits; completing a thir-teen item Likert-scale trait/state list (e.g., “I am happy,” “I feel good about my-self,” “I am angry,” “I do not play by the rules,” scored from “Almost Always” to “Rarely.” Those who scored “Almost Always” or “Usually” to positive states/traits, and “Not Often” or “Rarely” to negative state/straits were consider-ed free of clinical symptoms, at least by self-report. There were 200 women, ages 18 through 58, of varying educational and vocational backgrounds, and 82 men, ages 18 through 75, also of varying educational and vocational back- grounds, who fit the norm criteria. While African Americans were somewhat under-represented, the other ethnic groups making up the norm sample were proportion to the United States population as reported by the latest Census Bureau findings. Significant mean differences were found between women and men on several scales, necessitating separate norms. Despite these interest-ing differences, as noted below, additional data will be being collected to obtain a larger and diverse sample to create adequate norms.

Scoring System

Many content analysis scoring systems – specific to apperception tests, as well as verbal/text analysis in general – were reviewed (e.g., Roberts, 1997; Smith, 1992). Initially, specific scoring categories with rigorous empirical evi-dence of validity (e.g., n Intimacy, McAdams & Vailiant, 1982) were going to be transformed into algorithms for computer-based scoring. The advantage, in addition to speed, is the elimination of any inter-rater (or intra-rater) error. Thus, a goal was to ensure that each SAGAS™ submission would be scored precisely the same way.

Fortunately, in reviewing the literature, the current author discovered the Psychiatric Content Analysis and Diagnosis (PCAD) computer scoring program of written behavior developed by Louis Gottschalk, M.D., and Robert Bechtel, Ph.D. (Gottschalk & Bechtel, 1993). Based on the content analysis system developed by Dr. Gottschalk and his colleagues over the past thirty years, the program results in a highly detailed profile of the individual across numerous affective and personality states and traits. For example, there is an Anxiety scale, with subcategories such as Separation Anxiety and Guilt Anxiety; Hostility Directed Outward and Inward scales; Social Alienation and Personal Disorganization scale; Cognitive Impairment scale; Hope scale; Depression scale; and Health/Sickness scale, each with various subscales.

What is most impressive about the Gottschalk scales is the span of re-search supporting the validity of the various measures. There literally have been hundreds of studies, both in the United States and abroad, validating the scales, some even at the neuro-biological level (for summaries of studies see Gottschalk, 1979 and 1995; and Gottschalk and Gleser, 1969).

The SAGAS™ submissions lend themselves well to analysis by the PCAD program, creating a unique profile of the respondent along a variety of meaningful and previously normed dimensions. Indeed, the feedback from those so far who have received profiles is that the test is extremely telling – to quote a recent respondent, “The results were unbelievably accurate…I can not believe that you gleaned such specific details through one test.”

FUTURE DEVELOPMENT

The SAGAS™

The author is in the process of expanding the norms for the SAGAS™ as he continues to obtain submissions from those who fit the criteria. He also is very interested in obtaining profiles for specific clinical populations. For ex-ample, a colleague of the author, who works with adult perpetrators of child sexual abuse, will be having his clients take the test. Furthermore, various colleagues of the author who perform custody assessments will be utilizing the instrument with this population. Also, there is an open invitation for graduate students and others involved in research (including any reading this article), to use the SAGAS™ in their work – profiles will be supplied at no charge. Outcome studies and correlation studies with other instruments assessing a variety of psychological, cognitive and emotional variables certainly are necessary and fertile areas of inquiry. The transforming of existing apperceptive scoring systems (e.g., for n Intimacy) into computer programs is another area of development.

Other Tests

The author is in the process of norming an apperception test for children administered over the Internet by parents or others. The test – called the Parent Administered Child Test or Multi-person Administered Child Test (the PACT™ or M- PACT™) - utilizes seven pictures depicting a day in the life of a boy or girl (there is a separate set for African-American, Asian, Caucasian, Hispanic, and wheelchairbound girls and boys). Scoring systems similar to those of the Roberts Apperception Test for Children (McArthur & Robert, 1982) and the Children’s Apperceptive Story-Telling Test (Schneider, 1989) will be developed into a computer program.

The author also invites colleagues who would wish to join him in devel-oping other tests suitable for Internet administration. It is clear that the possibi-lities for assessment of adults and children, and the opportunity for research on a global level, are unprecedented and astounding indeed!