The Origins and Development of Frederic Kuder’s Preference Record

Origins

How did Kuder, then a new PhD in a nonteaching position at the University of Chicago, manage to develop such a novel and well-received interest assessment? He had earned a bachelor’s degree in English at the University of Arizona in 1925, after which, through his friendship there with one of James McKeen Cattell’s sons, he found employment in New York City as an editor for Cattell’s American Men of Science.

After 2 years in New York, Kuder decided to seek a master’s degree in vocational guidance and enrolled in the master’s degree program at the University of Michigan under George E. Myers, then the president of the newly formed National Vocational Guidance Association. Myers was preparing a book, Planning Your Future (Myers, Little, & Robinson, 1929), detailing a yearlong junior high school vocational exploration class, and including 10 chapters describing occupations in various sectors, such as agriculture, public service, clerical, and the like. Kuder said (Zytowski & Holmberg, 1988) that he “wrote a couple of these chapters.”

When Kuder finished his master’s degree, in 1929, employment opportunities were severely constricted, and he was pleased to find a job in the personnel research department of Procter and Gamble (P&G) Co. The department had been established 4 years earlier by Kenagy (1925), formerly of the Bureau of Sales Research of the disbanded Department of Applied Psychology at the Carnegie Technical Institute (later Carnegie Mellon University). Kenagy (1925) was the coauthor of the book, The Selection and Training of Salesmen, with Yoakum, who is remembered for his 1920 seminar that produced the 1,000-item interest inventory that several students used in their dissertations, and from which Strong drew inspiration. Kuder said (Zytowski & Holmberg, 1988, p. 153) referring to Strong’s inventory, which he had seen during his master’s degree studies, that it was while he was at P&G that he thought he could produce a better interest inventory.

In Kuder’s time at P&G, the department was headed by Uhrbrock, a 1928 Columbia University PhD who had earned his master’s degree in the Carnegie program under Strong, with whom he coauthored a book (Strong & Uhrbrock, 1923) on the printing industry. Uhrbrock’s main responsibility was the development and application of tests for sales personnel and supervisors. ¹ Kuder’s immediate supervisor was Marion Richardson, then a PhD candidate at the University of Chicago, and later a founding partner of the consulting firm of Richardson, Bellows, Henry & Co, of Arlington, VA. Uhrbrock and Richardson (1933) early collaborated on a test to forecast supervisory ability. One of Kuder’s first assignments was to evaluate 2,000 items for inclusion in this test.

Kuder’s first published research (Richardson & Kuder, 1940) recounts the construction of a scale measuring the effectiveness of P&G’s case-goods salesmen. They describe it as easy to administer and score, and showing reliability coefficients ranging from .83 to .90, replacing a notoriously unreliable graphic rating scale then in use. It is clear that Kuder, who had intended to become a school counselor, was introduced to tests and measurements at P&G by people intimately associated with the origins of interest measurement. Kuder’s Preference Record and Strong’s (1927) Vocational Interest Blank were rooted in the same soil, notwithstanding their several differences.

In 1934, at the depth of the Great Depression, P&G closed their personnel research department. Kuder decided to continue with PhD study at the Ohio State University under Herbert Toops, who was well known for his development of The Ohio State Psychological Examination, a college admissions test. Kuder completed his course work in 1937 and found employment in a nonteaching position at the University of Chicago, preparing standardized final exams that were also used for students to “test out” of class attendance for course credit.

The Preference Record Methodology

It was not until the manual for Form B of the Preference Record that Kuder (1942) recounted the experimental work that was its foundation. In his first year of doctoral study, he assembled 40 sets of five activities, each in the form of a transitive verb and a direct object (e.g., browse in a library; work in a garden). He submitted these to 500 Ohio State students, asking them to rank each set of five in the order of their preference. Each set of five was taken as 10 pair comparisons, yielding 400 paired items, from which could be calculated, in effect, covariances. From these data, he identified sets of items that covaried strongly and, when assembled into scales, calculated their reliabilities. One set of items showed an internal consistency of .85; Kuder named it the Literary scale. A second set of items that were correlated with each other at about .65 but not to the Literary scale formed the Experimental (later named Science) scale. Items that had low correlations with the Literary and Experimental scales but correlated with each other were assembled and named “Artistic.” Yet a fourth scale was drawn from this data set that Kuder thought reflected social prestige. Kuder assembled several more sets of activities, from which other scales were extracted. It was found that the prestige scale was too highly correlated with several of the new scales, and was dropped. Other scales were extracted and correlated with existing scales, resulting in the dropping of religion and athletics scales, but with the first developed scales comprised the seven scales of the Form A Preference Record: Literary, Scientific, Artistic, Musical, Computational, Social Service, and Persuasive. How Kuder (1938b) was able to calculate the many needed correlations is likely revealed in his article on the use of the IBM test scoring machine for the rapid calculation of tables of intercorrelations.

SRA

In May 1939, the University of Chicago Bookstore manager introduced Kuder to two graduate students in sociology, Bob Burns and Lyle Spencer, who in 1938 had formed a nonprofit company, SRA, to publish occupational information materials for sale to school counselors. In an interview (Grele, 1984), Kuder said that he was impressed with the enthusiasm of the young entrepreneurs and decided to give them publication rights to his Preference Record, despite interest from his former employer’s more-established Psychological Corporation. In 1940, SRA became publisher of the Kuder Preference Record, Form A. The slightly enhanced manual was letterset, but the test booklet and the several forms of answer sheets were the same as in the typewritten Bookstore edition. Kuder’s group at the University of Chicago having been dissolved, he moved to Washington, DC, to begin a series of jobs starting with the Civil Service Administration, and continuing in the War Department for the duration of World War II. He remained living in the DC area, and from an office on K Street, refining the Preference Record and editing Educational and Psychological Measurement, until in 1948 he joined the faculty of the Duke University.

Form B: The Preference Record Improved

A new edition of the Preference Record, Form B (Kuder, 1942) replaced Form A. It featured two additional scales, Mechanical and Clerical, suggested, Kuder says, by users of the earlier form. Form B’s test booklet contained 168 items, but to save administration time now presented in triads; one activity to be marked most-preferred, one least-preferred, and one left unmarked. Scoring gave the most-preferred item a weight of 2, the unmarked (next most-preferred) Item 1, and the least preferred 0. In addition to public sales, this form found use in several applications in the Armed Services during World War II, to which Kuder waived his royalties.

In response to requests by users, Kuder says, the manual included directions for calculating a masculinity–femininity score. Means and standard deviations for men and women in several occupations were given for this score, but there was no discussion of its use.

A 1946 manual for Form B presented mean score profiles for a number of male and female occupational groups contributed by users, suggesting that Kuder understood the appeal of an interest inventory that addressed occupational prospects. Numbers varied from as few as 16 English teachers to 195 meteorologists. Kuder only comments that “they indicate that the names assigned to the scales seem appropriate in terms of the occupation entered.”

The Form B Kuder attracted considerable professional attention. Buros’ (1949) Fourth MMY listed 62 references. Baker and Peatman’s (1947) study of test usage by Veteran’s Advisement Centers revealed that the inventory was used in 99% of the centers, compared to 70% for the Strong Blank, and 35% or less for any of the remaining interest inventories available. Reviewer Ralph Berdie noted that Kuder gave scale scores for men and women in 72 different occupations, while Strong gave data for 56, although the number of individuals in Strong’s groups tended to be larger than Kuder’s. Donald Super’s review suggested that the Preference Record might be more valuable than the Strong for use with women, owing to its better differentiation.

Form C: The Preference Record Perfected

Form C of the Preference Record was issued in 1948, with one additional scale, Outdoor, indicating a preference for work that “keeps one outdoors most of the time, usually dealing with animals and growing things.” The manual for this form suggests that in addition to vocational counseling, the Preference Record could be valuable in employee screening and placement, as well as provide help in identifying material that might be especially interesting to people seeking to improve their reading skills (SRA having introduced a substantial reading instruction program at the time).

Form C was notable for introducing the Verification (V-) Scale. It consisted of 45 items to which almost everyone responded either Most or Least, so that anyone responding carelessly might endorse many items that are rarely preferred or fail to endorse items that are almost always preferred. It indicates in two directions: A score higher than 45 suggests that the respondent has checked more than one Most and one Least per triad, while a score below 37 suggests that the survey taker had failed to respond to all items or has responded insincerely, or could be a poor reader. Occasionally, people reared in a non–American or western culture score low on the V-Scale, despite their assurance that they had responded sincerely. Campbell (1965) praised the V-score in his Buros review, and his revision of the Strong Blank for men (Campbell, 1965) included several administrative indices, one of which consisted of infrequently endorsed items.

Form C’s manual was revised frequently in order to report empirically derived occupational profiles in the manual, relieving, in part, the responsibility of the test giver to generate occupational prospects from the respondent’s profile. For example, one edition provided mean scale scores for 581 accountants, but as few as 27 actors. The 75th percentile was suggested as a significant high score: 111 psychologists’ highest mean scores were 84 on the Science and Literature scales, while electrical engineers averaged 85 on Science and sales engineers averaged 83 on Persuasive and 57 on Science. Among women, 22 art teachers’ highest mean score was 97 on Art and 70 on Mechanical.

The fourth MMY (Buros, 1953) review of Kuder’s Preference Record listed 208 references, attesting to the growing popularity of the inventory. Ralph Berdie concluded his review by stating “that the Preference Record has been found clinically useful in a number of applications” and “that it is easy to administer, only slightly monotonous to the individual tested, and convenient to score.”

Form C’s Legacy

In 1948, Kuder introduced the Personal Preference Record, Form A. He described it as a supplement to the vocational form to help determine “the kind of situation in which the person prefers to work.” It reported five scales: (A) Sociable: preference for participating in groups, (B) Practical: preference for familiar or stable situations, (C) Theoretical: preference for working with ideas rather than things, (D) Agreeable: preference for avoiding conflict, and (E) Dominant: preference for directing or influencing others. Kuder noted that there are important differences in personal preferences within occupations characterized by certain high scores on the vocational form. For instance, while accountants may be generally characterized by their interest in computational tasks, some accounting specialties may be differentiated by their preferences for avoiding conflict or for directing others. It was reviewed as a nonprojective personality assessment in Buros’ (1959) fifth and subsequent editions of the Mental Measurements Yearbooks.

Kuder’s General Interest Survey, Form E, released in 1953, was a revision of Form C designed for use with younger people by reducing the reading level and adding more items to support scale reliabilities.

Kuder (1956) introduced the Occupational Interest Survey, Form D, presenting Form C’s items, but scored, like Strong’s Blank, for occupations. It differed from the Strong in foregoing the use of a general reference group, affording male- and female-normed scales to be reported on the same report, irrespective of the gender of the inventory taker. The manual described Kuder’s method of constructing an occupational scale, so that users could generate their own scale for any occupation for which they could recruit a criterion group.

Campbell (1965), who succeeded Strong in 1963 as developer of the Strong Blank, provided the review of Form D in the sixth MMY (Buros, 1965). He praised the concurrent validity of Form D, but was critical of the lack of predictive and construct validities. He noted Kuder’s Verification scale as a worthwhile feature, and found that the Kuder “has the edge over Strong” in being shorter and having simpler scoring, but advises practitioners to continue with the Strong.

Just as the 1969 Strong Blank added Holland’s (1979) homogeneous interest scales, Kuder’s (1966) Form DD merged Form C’s vocational interest scales with Form D’s occupational scales to inform inventory takers of their activity preference profile and the occupations with which these interests were most consistent. The report form gave occupational scale scores in one table for both male and female occupational criterion groups in rank order.

Kuder (1980), observing the varying homogeneities of interests within occupations, abandoned occupational scaling entirely and introduced his concept of Person Match. He replaced criterion groups with an array of “criterion persons,” consisting of successfully employed individuals, accompanied by narrative accounts of their work histories. Person Match is currently a key component of the Kuder Career Interests Assessments in the Kuder Career Planning System. A full description of the development of Forms D and DD and Person-Match is given in Zytowski (1992).

Borgen (1986), in his account of the diverse approaches to the measurement of interests, dubbed Kuder’s Preference Records, Strong’s Blank, and Holland’s (1979) Self-Directed Search the “Big Three” of interest measurement. While the Occupational Interest Survey ceased publication in the late 1990s, the Form C Kuder Preference Record, with periodically updated norms, remained in print until 2002—more than six decades of service to millions of Americans and others in translation worldwide.