Employee job-satisfaction measurement developed from broad, often informal judgments about morale into standardized, multidimensional psychometric instruments. The change was tied to a wider shift in industrial and organizational (I–O) psychology: from treating productivity chiefly as a function of physical conditions and pay toward studying workers’ perceptions, social relations, supervision, and the meaning of work.

1. From scientific management to the human-relations problem

Early industrial psychology inherited a scientific-management emphasis on training, task conditions, equipment, and efficiency. The Hawthorne program at Western Electric began in the 1920s with experiments on illumination and working conditions, but its later stages made employee attitudes a central object of inquiry. Between 1928 and 1931, researchers interviewed about 21,000 workers; the interviews were first relatively direct and later became nondirective, allowing employees to choose the topics they considered important. The records were used for supervisory training, workplace improvements, and research on satisfaction and dissatisfaction (Kompier, 2006).

The methodological lesson was that satisfaction could not be inferred from physical conditions alone. The Hawthorne interviews found that attitudes were shaped by supervision, work groups, home life, and community circumstances, and that the same situation could be satisfying for one worker and dissatisfying for another (Hawthorne Experiments, n.d.). This widened the measurement problem: researchers needed to capture subjective evaluations, not merely observe output or manipulate the workplace. The later historical account of the Hawthorne studies describes their influence as a stimulus for hundreds of studies linking employee attitudes with performance, while also noting their methodological weaknesses (1925–1950, n.d.).

2. Attitude measurement enters industrial research

The technical foundation came from general attitude measurement. Louis Thurstone’s work in the late 1920s defined attitudes in terms of the direction and intensity of affect toward an object and demonstrated that attitudes could be scaled. Thomas Wright’s historical review reports that Kornhauser’s 1933 applied work adopted Thurstone’s framework for employee attitudes and that Uhrbrock’s 1934 research used a purpose-built scale in a large manufacturing-firm study of 4,430 employees (Wright, 2006).

Rensis Likert’s 1932 dissertation, A Technique for the Measurement of Attitudes, simplified Thurstone’s more laborious scaling procedure. Respondents rated statements on ordered agreement categories, and the responses were summed; a historical account reports that Likert’s simpler method produced results comparable to Thurstone’s procedure with fewer items (The Origin and Development of Rating, n.d.). This format made anonymous questionnaires practical for organizational use. The historical review of American employee-attitude testing places the emergence of worker surveys in the 1920s, their widespread use in the late 1930s and after World War II, and their purposes in improving employee relations and loyalty. It also records a limitation that shaped the field: inexperienced survey administrators encouraged industry to form closer ties with academic behavioral scientists (Jacoby, 1988).

The survey was therefore never only a neutral measurement device. It served both a scientific purpose—quantifying otherwise diffuse attitudes—and a managerial purpose—diagnosing morale, supervision, communication, and possible sources of conflict. The Sears program, for example, became an especially sophisticated corporate application of behavioral science during the 1940s and 1950s, although it was also connected to the company’s effort to forestall unionization (Jacoby, 1986).

3. Robert Hoppock and the measurement of general satisfaction

Robert Hoppock’s Job Satisfaction (1935) was an important bridge between attitude scaling and job-satisfaction research. A later measurement history identifies Hoppock’s instrument as the first contemporary job-satisfaction measure and describes it as a four-item measure of general satisfaction (Development of a Compact Measure of, 2026). Hoppock’s New Hope, Pennsylvania survey obtained responses from 88% of 351 employed adults; the reported results classified about 15% as dissatisfied, while a later summary of the same line of work reports that only 12% were classified as dissatisfied. The discrepancy reflects differences in the retrieved summaries, so the defensible conclusion is that dissatisfaction was a minority position in that sample (Measuring Job Satisfaction (With Diagram), 2015) (11-2, n.d.).

Hoppock’s contribution was broader than a single score. His design treated satisfaction as a worker’s overall evaluation of employment and compared people across occupations and personal circumstances. The retrieved historical account reports higher satisfaction among professionals and managers than among unskilled manual workers and substantial variation within occupational groups (11-2, n.d.). Hoppock also extended the program to teachers: a summary of his work describes four attitude scales administered to 500 teachers in 51 communities, followed by a much larger set of questions for the most- and least-satisfied groups (Measuring Job Satisfaction (With Diagram), 2015).

He then helped create a cumulative research tradition. Hoppock and H. Alan Robinson published recurring reviews of job-satisfaction research, including reports for 1948, 1949, 1950, and 1951, rather than treating the 1935 study as an isolated finding (Hoppock, 1949) (Hoppock, 1950) (Hoppock, 1951) (Robinson, 1952). His 1937 study of psychologists illustrates the portability of the approach: a job-satisfaction questionnaire was mailed to 203 psychologists and answered by 66, whose average self-rating placed them at the 64th percentile of the employed-adult population (Job satisfaction of psychologists, n.d.).

4. From a global score to job facets

By the 1950s and 1960s, researchers increasingly distinguished overall satisfaction from satisfaction with particular job facets. The logic was practical and theoretical: pay, promotion, supervision, coworkers, security, autonomy, and the work itself could contribute differently to a person’s global evaluation. The historical literature describes a movement in which questionnaires became the primary method, facet measurement became the primary content, and organizational effectiveness became a major rationale for studying attitudes (Job Satisfaction, n.d.).

Two influential systems consolidated this approach in the 1960s. Smith, Kendall, and Hulin’s Job Descriptive Index (JDI), published in 1969, measured five facets with descriptive items concerning the work, pay, promotion, supervision, and coworkers. The Minnesota Satisfaction Questionnaire (MSQ), associated with Weiss, Dawis, England, and Lofquist’s work, provided a broader facet framework and long and short forms; later summaries describe it as assessing 20 facets (Development of a Compact Measure of, 2026) (White, 2008). The JDI’s later development also shows the field’s continuing psychometric concern: a 72-item version was reduced to a 25-item abridged instrument while preserving relationships among the facet scores and with other measures (Development of a Compact Measure of, 2026).

This facet tradition corrected a weakness in single global questions. An employee could be satisfied with the work itself and coworkers but dissatisfied with pay or promotion. Measuring facets made it possible to distinguish these profiles instead of collapsing them into one overall number (What is employee satisfaction?, 2018). It also made comparisons across jobs and organizations more informative, provided that the same constructs and scoring rules were used.

5. Psychometric expansion and continuing problems

The dominant modern format became the summated-rating or Likert questionnaire: respondents rate statements from strong disagreement to strong agreement, and item scores are combined into a total or subscale score (How Job Satisfaction Is Measured, 2023). Instruments diversified into global measures, facet measures, visual formats such as the Faces Scale, and job-characteristics measures such as the Job Diagnostic Survey (11-2, n.d.). Later work also tested whether instruments function equivalently across groups. For example, Swander’s analysis of a polytomous employee-attitude survey examined differential item and test functioning for managers versus nonmanagers and for medical versus nursing staff (Bliese, 2017).

The field’s central unresolved issue has been standardization and construct validity. A review of measurement practices found that 71 measures—46% of 155 instances in which job satisfaction was measured in five leading journals during 1973–1975—appeared only once; the authors argued that this limited replication and cumulative knowledge (O’Connor, 1978). The history therefore has a tension at its core: surveys became easier to administer and more statistically sophisticated, but the proliferation of locally created questionnaires made results difficult to compare.

Synthesis

The historical sequence is best understood as four linked developments: Hawthorne-era observation and interviewing made worker attitudes organizationally consequential; Thurstone and Likert made attitudes quantifiable; Hoppock demonstrated that general job satisfaction could be surveyed systematically across workers and occupations; and the JDI, MSQ, and related instruments decomposed satisfaction into measurable facets. Contemporary practice retains all four legacies: anonymous self-report, Likert-type response formats, separate global and facet scores, and psychometric tests of reliability, validity, and group comparability. The evidence also cautions against reading a single satisfaction score as a complete account of work experience: the historical studies repeatedly found that satisfaction varies across people, job aspects, social settings, and organizational conditions (11-2, n.d.) (Hawthorne Experiments, n.d.).

The accessible historical material is stronger on the American development of I–O measurement than on parallel early traditions outside the United States; the account above should therefore be read primarily as a history of U.S.-centered industrial psychology and employee-attitude surveying.