College and Research Libraries Job Satisfaction among Library Support Staff in Alabama Academic Libraries Donna K. Fitch The study emphasizes library support staff, a largely neglected group, and discusses a survey of the job satisfaction of these employees in academic libraries in the state of Alabama. The Job Descriptive Index (JDI) was used as the survey instrument, and the resulting satisfaction scores were related to the variables of size of city, sex, variability of working hours, type and size of institution, staff, department, historic race of the institution, and automation status of the library's functions. The null hypothesis could not be rejected for most variables, reaffirming conclusions of other studies that demographic factors do not influence job satisfaction. The need for better compensation and opportunities for promotion for academic library support staff is emphasized, as well as the need to relate satisfaction to quality of work-life issues. ob satisfaction is an area that has been extensively studied in the business world since the 1930s, yet forty years passed before library-oriented studies began. The majority of these studies have focused on professional and paraprofessional librari- ans, and thus the needs and attitudes of li- brary support staff have been largely over- . looked. The present study explores some factors that may contribute to job satisfac- tion among support staff in academic li- braries. More than 6,000 articles on job satisfac- tion had been written by 1984, prompting the question: why embark on another study?1 As Patricia Cain Smith, Lorne M. Kendall, and Charles L. Hulin point out, little evidence exists successfully linking job satisfaction and productivity, so that frequently discussed topic is hard to sup- port as a valid reason for additional re- search.2 However, Beverly P. Lynch and JoAnn Verdin, in their study of job satis- faction in libraries, indicated a need for more studies that would be ''conducted within the framework of the work itself. '' 3 Few studies have been conducted in li- braries, and many unexplored facets of job satisfaction in this context remain. Aside from purely scientific reasons for such research, there is a humanitarian rea- son as well. Employees who must spend eight hours a day at a job should enjoy what they are doing, as Smith, Kendall, and Hulin, and Susanne Wahba indi- cate.4'5 Studies in improving the quality of working life, such as those indicated in Charles Martell's 1981 article, have pointed out the need for redesign of work systems. 6 This redesign involves an atten- tion to the needs of the employee, as well as the needs of the organization. Al- though the factors explored in the present study cannot be controlled by library man- agers, knowledge that dissatisfaction ex- ists can assist further studies in disclosing and improving problem areas within the organization. Attention needs to be paid to satisfaction in connection with the ten- Donna K. Fitch is a Catalog Librarian at the University Library, Samford University, Birmingham, Alabama 35229. 313 314 College & Research Libraries ets of work system design, especially since many libraries are reorganizing in re- sponse to computer technologies. Attention needs to be paid to satisfac- tion in connection with the tenets of work system design, especially since many libraries are reorganizing in re- sponse to computer technologies. Interest is added to the results by the lo- cation in which the study was conducted, and its possible relation to job satisfaction. In 1987, Alabama had a personal income per capita of $10,673, ranking forty-fifth among the fifty states, as compared with the U.S. average of $13,876. 7 DEFINITION OF TERMS Before continuing with the discussion, a few definitions are in order. The first is 11 job satisfaction'' itself. The subject of job satisfaction is complex and has been de- fined in many ways. One author defines it as "the individual's emotional reactions to a particular job," while another refers to it as 11 the feeling an employee has about his pay, his work, his promotion opportU- nities, his co-workers, and his supervi- sor.' ' 8' 9 Michael Beer defines job satisfac- tion as 11 the attitude of workers toward the company, their job, their fellow work- ers and other psychological objects in the work environment." He adds that "[a] fa- vorable attitude toward these indicates job satisfaction and vice versa.' ' 10 Smith, Ken- dall, and Hulin define job satisfaction as 11 the feelings a worker has about his job. " 11 Another phrase frequently used in this study is "support staff," meaning library employees whose educational level does not include the master's in library science (M.L.S.), and whose positions support the functioning of the M.L.S.-degree li- brarians. Included in the definition are clerical and paraprofessional employees. Job satisfaction has been frequently de- July 1990 bated, discussed, and researched since Robert Hoppock wrote Job Satisfaction in 1935. u Most of the literature has coalesced around several major theories, including Maslow's need hierarchy theory, Herz- berg's dual-factor or motivation-hygiene theory, and the theory of work adjust- ment. Readers interested in the general lit- erature of job satisfaction are referred to the bibliography compiled and edited by Ruth M. Walsh and Stanley Birken, Job Satisfaction and Motivation: An Annotated Bibliography. 13 In addition to general research on job satisfaction, more specific factors have been studied to determine their effects. As in this study, age, sex, and environmental characteristics have. been explored, as well as community characteristics, organiza- tional size, and geographical factors. Some of the references detailing these specific factors are mentioned throughout this article. Satisfaction studies in libraries have been few, and those studying non- M.L.S.-degree staff are almost nonexis- tent. One study by Lynch and Verdin used all full-time library employees as subjects, and dealt with the factors of the work itself, age, sex, occupational group, t,enure, supervisory level, career commit- ment, and library department. 14 They tested the null hypotheses of the relation- ships between these factors and job satis- faction, and found that: (1) job satisfaction was unrelated to sex; (2) older employees were more satisfied than younger work- ers; (3) people with more experience tend to be more satisfied; ( 4) those planning to stay in the same library are more satisfied; (5) nonsupervisors have lower satisfaction than supervisors; (6) reference depart- ment employees are more satisfied than any other department except acquisitions; and (7) professionals were more satisfied than library support staff members. This last finding contrasts with Lawrence Pry- bil' s report of no significant difference among occupational levels. 15 Asadollah Azad' s 1984 dissertation explored the job satisfaction of paraprofessionals, primar- ily comparing satisfactions in public and technical services departments. 16 STUDY DESIGN For the present study, the Job Descrip- tive Index (JDI), developed by Smith, Kendall, and Hulin, was used as the in- strument. The Index consists of six scales, Work on Present Job, Present Pay, Opportuni- ties for Promotion, Supervision, Co-workers, and Job in General. The scores for these scales are not combined, although many researchers have done so. The authors maintain that the individual factors are not equally weighted and cannot be satis- factorily combined. 17 Each scale consists of eighteen ques- tions, except for Present Pay and Promo- tions, which have nine. Responses are yes, no, or a question mark indicating indeci- sion. Validity of the JDI has been estab- lished through various trials, details of which can be found in Smith, Kendall and Hulin's The Measurement of Satisfaction in Work and Retirement. The prospective institutions for the present study were chosen from member libraries of the Network of Alabama Aca- demic Libraries, a consortium of public and private four-year college and univer- sity libraries in the state. Directors of fif- teen of the seventeen libraries permitted their employees to participate and pro- vided the investigator with lists of their li- brary support staff. From these lists, a stratified random sample of 185 subjects was chosen, and copies of the JDI were sent to them. Included in the packet was a series of demographic questions related to the variables, a letter explaining the sur- vey, and an envelope for return of the in- strument. To maintain demographic sta- tistics on town size, etc., a code number was written on the survey. Interestingly, despite assurances in the cover letter of anonymity, several respondents from one institution cut off or obscured the code number upon returning the instrument. Information about size of the town, size of the institution, size of the library's staff, whether public or privately owned, and the library's automation status was deter- mined from the American Library Direc- tory. 18 The remaining replies came from the subjects themselves. Job Satisfaction 315 The study dealt with seventeen vari- ables and their effect on job satisfaction. The null hypothesis was used in analyzing the variables. These factors can be divided into several groups, as follows: Demographic variables (size of the town or city in which the institution is located); Characteristics of the institution (public or private, its size, if historically black or white); Characteristics of the library (size of the library staff, if the library's functions were automated or not); The subject (sex, age, educational level, income, number of years in library work); and The job itself (full- or part-time, day- or nighttime, fixed or variable hours, depart- ment in which the employee works, de- partment size, amount of time spent working on a computer terminal). RESULTS The return rate was 64 percent, with subjects from sixteen libraries participat- ing, including separate law or medical li- braries from some institutions. The sur- veys were scored, and the data entered using SYSTAT-the System for Statistics, published by SYSTAT, Inc. 19 In these instances, the null hypothesis can be rejected; in all others, the null hy- pothesis cannot be rejected. These results are similar to those in Steven Seokho Chwe's 1978 study, which also found that demographic variables have no effect on job satisfaction. 20 The size of the institution was found to have a significant effect on satisfaction with supervision (P < .05), but not on the other scales. Employees in libraries in small schools (less than 3,000 students en- rolled) were most satisfied with supervi- sion, while those in medium-sized institu- tions (3,000-9,000) were least satisfied. Beer, in his article "Organizational Size and Job Satisfaction," cites interviews conducted by James Worthy in 1950 which ''indicated that organizational size was the single-most important variable re- sponsible for low job satisfaction.''21 Wor- thy went on to state that "morale and job satisfaction are related to integration (co- 316 College & Research Libraries hesiveness) and that integration is related to structural comElexity and in turn orga- nizational size.'' Other studies cited by Beer indicate the same principle. A Morse and Reimer study implies that ''decision- making levels become more and more re- mote as the organization becomes larger.' ' 23 Medium-sized schools, in many. cases, are in a state of transition from small to large. In small schools, more socialization takes place; people are able to learn more about each other and to become more fa- miliar with their supervisors. In larger schools it may be obvious that contact with supervisors cannot be as personal. But in transitional organizations, workers who were employed at the library when it was smaller may feel that they have less contact with their supervisors than they formerly did. This new situation leads to frustration and feelings of being left out or slighted. In addition, schools of this size may be implementing or planning to implement automation, a change that has a major ef- fect on employees. In their book on man- aging organizational change, Patrick E. Connor and Linda K. Lake quote John Ad- ams: "All changes are irksome to the hu- man mind, especially those which are at- tended with great dangers and uncertain effects.' ' 24 While the advent of automation cannot be said to be a great danger (al- though some might even argue with that), it does have "uncertain effects" for those whose jobs are involved. Lester Coch and John R. P. French, Jr. found that "resistance to change is a com- bination of an individual reaction to frus- tration with strong group~induced forces. " 25 They found that group resis- tance to change can be reduced or elimi- . nated through the use of communication in group meetings. 26 This idea is repeated by Martell, who said that ''information should be available when and where it is needed. Traditionally, management has hoarded information and distributed it only when necessary to maintain its source of power. This practice needs to be severely limited.' ' 27 Better communication may solve, or improve, many dissatisfac- . tions in medium-sized organizations. The institution of work groups instead of the July 1990 traditional division of labor frequently used at clerical levels in libraries may be another answer, enhancing socialization as well as providing employees with more responsibility. The way the question on supervision is asked in the Job Descriptive Index makes it difficult to determine what level of su- pervision is being addressed in terms of satisfaction or dissatisfaction. The state- ment in the JDI instructions says ''Think of the kind of supervision you get on your job," but does not specify which level. Women were more satisfied on all the scales than men. In this study, there were more signifi- cant differences in sex than in any other variable. Women were more satisfied on all the scales than men. Men's norms were an average of 14.01lower than JDI norms, while women's were an average of 6.17 lower. The lowest difference in both men's and women's scores carne in the area of Present Pay, 19.19 points lower than the JDI norms for men, 12.53 lower for women. This bears out H. Jack Shapiro and Louis W. Stern's finding that "non- professional women are more satisfied with their pay than are non-professional men.'' 28 Michael J. Kavanagh and Michael Halpern studied university employees and compared the job satisfaction and life satisfaction of men and women at profes- sional and nonprofessional levels. They compared their results to a 1957 study by Brayfield and Wells, which found ''no sig- nificant relationships between life and job satisfaction for females. " 29 In the 1973 study, however, strong relationships were found, especially at job level one (nonsupervisory, nonprofessional, or clerical). They theorize that ''the organiza- tional climate and attitudes found in a uni- versity environment would be _more sup- portive of the women's movement than those found in many other organiza- tions."30 The 1975 study by Shapiro and Stem also looked at job satisfaction of males and females at both the professional and non-. professionallevels. 31 In the nonprofes- sional sample, which included clerical workers, men were more satisfied with the work itself than women, while the women were more satisfied with supervi- sion. A 1975 article by W ahba discusses job satisfaction in terms of need fulfillment, need deficiency, and perceived need im- portance. She found that II women have significantly higher deficiencies than men in four areas: security, autonomy, esteem, and self-actualization needs," areas that were defined in the works of Maslow. 32 W ahba explains this difference by saying that women have a I' stronger desire or ex- pectation ... for a higher degree of secu- rity'' than men. 33 She relates the need ful- fillment score to job satisfaction, and the need deficiency score to both the 'I per- sonal expectations of need fulfillment and the actual fulfillment from the job. " 34 W ahba' s study dealt with professional li- brarians. Lynch and Verdin, in a study that in- cluded all full-time employees, concluded that there is no significant difference be- tween male and female employees in job satisfaction. 35 George P. D'Elia reached the same conclusion. 36 However, Smith, Kendall, and Hulin uncovered substantial differences in male and female norms, ex- plaining that "women are less satisfied overall than men because they receive less with which to be satisfied,'' but that ' 1 with a comparable level of income, women are more satisfied than men." They theorize that the reason is possibly because of dif- ferent frames of reference. 37 Does the answer lie in the fact that only 14 percent of the respondents were men? Although attitudes are changing, as the references cited above suggest, older ideas about the "proper place" for men and women may still exist in a region such as Alabama, which still adheres to tradi- tional values that consider the male the source of the family paycheck. In a low- paying profession with little opportunity for advancement, possibly still viewed by some as "women's work," men become discontented. A person who is the sole support of his or her family cannot afford to work at the support staff level in a li- Job Satisfaction 317 brary. This may be the reason the majority of library workers in the state are female. In the area of income, there was no sig- nificance, except in the scale of Opportuni- ties for Promotion. Employees earning less than $8,000 per year (mostly part-timers) were least satisfied with their opportuni- ties for promotion, while those in the $9,000-$9,999 range were most satisfied. Part-time personnel generally have fewer opportunities to be promoted. Reduced working hours limit organizational oppor- tunity. The data gathered does not explain why those in the $9,000 range were most satisfied. Most of the scales in the "years worked in the library'' variable were not signifi- cant, with the exception of Opportunities for Promotion (P < .05). The most satisfied in this area, in partial contrast with the results of Lynch and Verdin's study, were those who have worked less than one year at the library; the least satisfied were the employees with ten or more years of ser- vice.38 At the support staff level, little op- portunity for advancement usually exists. The study by Kavanagh and Halpern found that as job levels increase, job satis- faction decreases. 39 One explanation they give for this seemingly contradictory find- ing is that stress rises as job levels in- crease. In this case, stress may also rise when job levels stop increasing, bringing about frustration and lower morale. As figure 3 shows, fully 24% of the li- brary workers surveyed earned less than the Alabama per capita income listed in the Statistical Abstract in 1987; about 75% earned less than the U.S. per capita in- come. Scores on the Present Pay scale for this survey were much lower than the JDI norms, as mentioned earlier, and the rea- son is evident. These results are markedly different from the ten-year study of job satisfaction in industry that reported higher scores in the South than any other • 40 regiOn. Most employees in the survey worked only in the library in which they were presently employed. Only about a quarter of them had worked elsewhere, suggest- ing that they view their library employ- ment as a job, not a career, and that they do not seek to move to other libraries as professional librarians do. One-third of 318 College & Research Libraries Sex: female: BG% male: 14% Median salary range: $10,000 to $14,999 per year Median number of years worked in that particular job: 4-G Subjects having worked in libraries other than the one in which she/he was presently employed: 25% Median age: 30-39 Average highest educational level attained: some college courses without completion of a degree. FIGURE 1 Demographic Results 957. probability level Opportunities for Promotion and income Work on Present Job and sex Supervision and size of the institution 997. probability level Opportunities for Promotion and years in the library Supervision and sex Job in General and sex FIGURE2 Significance of Dependent ana Independent Variables these employees have worked in their par- ticular job and library between one and three years. Whether their work-related plans include subsequent library work is unknown. Employees with less than one year's service are the most satisfied with opportunities for promotion, while those with more than ten years are least satis- fied. As mentioned previously, employ- ees are limited in how far they can ad- vance without an M.L.S. degree. One area in which no relationship with July 1990 Salary $25,000+ $15,000-24,999 $10,000-14,999 $ 7,000- 9,999 $ 5,000- G,999 $ 0- 4,999 Percentage 0% 21% 55% 20% 0% 4% .FIGURE3 Percentage of Respondents in Salary Ranges satisfaction was found was time spent on a computer terminal, probably because the question on the survey was not dear enough. The form of the question was ''What percentage of your present job is spent at a computer terminal?" Note that it did not specify day, week, or month. Employees with less than one year's service are the most satisfied with op- portunities for promotion, while those with more than ten years are least satisfied. The effect of video display terminal radia- tion on employees has been a frequently discussed topic in the past few years, but these articles are primarily concerned with health effects. They do not address the question of satisfaction, except in relation to job stress. An article by Robert I. Sutton and Anat Rafaeli in 1987 suggests that working at a terminal is a more complex is- sue than just measuring satisfaction and the percentage of time spent at that termi- nal. Their research found that for clerical employees, "characteristics of work sta- tions may not be occupational stres- sors. " 41 They discuss intrusions, noise, and heat as factors to be accounted for in determining satisfaction with work sta- tions. A 1981 article also shows that VDT- workstation-related stress· is a complex topic. 42 The need for further research in this area increases as library automation becomes more widespread. CONCLUSION The most urgent problems brought up by this survey are compensation and op- portunities for promotion. Money is tight for education in Alabama, as it is in many states. While immediate supervisors have little, if any, control over salaries, they can make the department a better place in which to work by keeping these findings in mind. Supervisors can encourage em- ployees who wish to pursue a degree or complete one, ensuring that they receive time off for such activities. Supervisors can be supportive of new ideas for im- proving the work place. Although not a panacea, positive attitudes can go a long way toward helping increase job satisfac- tion. In organizations that are undergoing change, particularly those automating the library, the communication process needs to be examined. Quality of work-life is- sues must be addressed when reorganiza- tion takes place, if managers wish to keep dissatisfaction to a minimum. The need is great for librarians to become familiar with principles dealing with quality of work- life and management of organizational change. Martell points out that organiza- tions, including libraries, are at an "ele- mentary stage in their use of modern orga- nization design techniques,'' and that the ''perceptual and technical skills required to develop and implement contemporary work system design may not yet exist within librarianship. ''43 Job Satisfaction 319 A study with as many variables as the present one naturally generates more questions than it answers. Would a repeti- tion of the study with a larger sample yield similar results? Studies performed in other states would also be enlightening. More information is needed about the work attitudes of the library support staff workers. Do they view their library em- ployment simply as a job, rather than as a career? Do they see their library jobs as "women's work"? How do the male em- ployees feel about their roles in the li- brary? Do the jobs males have differ from the ones females do? Job satisfaction in libraries continues to be a rich area for study, and each investi- gation generates further areas for study. The fact that most of the null hypotheses could not be rejected does not invalidate the study. The literature of library satisfac- tion studies is enriched by each bit of knowledge added to it, and results of some past studies are confirmed by the present study. The satisfactions of aca- demic library support staff in one of the poorer states have been explored; in gen- eral, they correlate with norms estab- lished for the Job Descriptive Index used in the survey. Library managers at all lev- els who wish to be successful should ex- amine their library in the context of job sat- isfaction studies, as well as quality of work-life and change management princi- ples, to discover what changes would help raise the level of satisfaction of the very human people who work for them. REFERENCES AND NOTES 1. G. E. Wittingslow and B. Mitcheson, "Job Satisfaction among Library Staff," Journal of Library Administration 5:61-69 (1984). 2. Patricia Cain Smith, Lome M. Kendall, and Charles L. Hulin, The Measurement of Satisfaction in Work and Retirement (Chicago: Rand McNally, 1969) . 3. Beverly P. Lynch and JoAnn Verdin, "Job Satisfaction in Libraries: Relationships of the Work Itself, Age, Sex, Occupational Group, Tenure, Supervisory Level, Career Commitment, and Li- brary Department," Library Quarterly 53:434-47 (1983). 4. Smith, Kendall, and Hulin, Measurement of Satisfaction, p.3. 5. Susanne Patterson Wahba, "Job Satisfaction of Librarians : A Comparison between Men and Women, College & Research Libraries 36:45-51 (1975). 6. Charles Martell, "Improving the Effectiveness of Libraries through Improvements in the Quality of Working Life," College & Research Libraries 42:435-46 (1981). 7. United States . Dept. of Commerce . Bureau of the Census. Statistical Abstract of the United States 1987 (Washington, D.C.: Govt. Print. Off ., 1986). 320 College & Research Libraries July 1990 8. M. M. Gruneberg, Understanding Job Satisfaction (New York: Wiley, 1979), p.3. 9. William J. Vaughn and J. D. Dunn, "A Study of Job Satisfaction in Six University Libraries," Col- lege & Research Libraries, 35:163-77 (1974), p .163. 10. Michael Beer, "Organizational Size and Job Satisfaction," Academy of Management Journal7:34-44 (1964), p.34. 11. Smith, Kendall, and Hulin, Management of Satisfaction, p.6 . 12. Robert Hoppock, Job Satisfaction (New York: Harper and Row, 1935). 13. Ruth M. Walsh and Stanley Birken, Job Satisfaction and Motivation: An Annotated Bibliography (West- port, Conn.: Greenwood Press, 1979). 14. Lynch and Verdin, "Job Satisfaction in Libraries," p.434-47. 15. Lawrence D. Prybil, "Job Satisfaction in Relation to Job Performance and Occupational Level," Personnel Journal52:94-100 (1973). 16. Asadollah Azad, "Job Satisfaction of Paraprofessional Librarians: A Comparative Study of Public and Technical Services Departments in Selected University Libraries," DAI 39:6379A (1984). Uni- versity of Pittsburgh. 17. Smith, Kendall, and Hulin, Measurement of Satisfaction, p.17 -18. 18. R. R. Bowker Co., American Library Directory, 2 vols. (New York: R. R. Bowker, 1987). 19. Leland Wilkinson, SYST AT -the System for Statistics, computer software (Evanston, Ill.: SYST AT, Inc., 1984). IBM-PC, 256KB, MS-DOS 2.0 or higher, disk . 20. Steven Seokho Chwe, "A Comparative Study of Job Satisfaction: Catalogers and Reference Li- brarians in University Libraries," Journal of Academic Librarianship 4:139-43 (1978). 21. Beer, "Organizational Size and Job Satisfaction," p.39. 22. Ibid . 23. Ibid., p.40. 24. Patrick E. Connor and Linda K. Lake, Managing Organizational Change (New York: Praeger, 1988), xiii. 25. Lester Coch and John R. P . French, Jr., "Overcoming Resistance to Change," Human Relations 1:520 (1948). 26. Ibid., p.531. 27. Martell, "Quality of Working Life," p.441 . 28 . H. Jack Shapiro and Louis W. Stern, "Job Satisfaction: Male and Female, Professional and Non- Professional," Personnel Journal 54:389 (July 1975). 29. Michael J. Kavanagh and Michael Halpern, "The Impact of Job Level and Sex Differences on the Relationship between Life and Job Satisfaction," Academy of Management Journal 20:69 (March 1977). 30. Ibid., p.71. 31. Shapiro and Stern, p.389. 32. Wahba, "Job Satisfaction of Librarians," p.48. 33. Ibid. 34. Ibid ., p.50 . 35. Lynch and Verdin, "Job Satisfaction in Libraries." 36. George P. D'Elia, "The Determinants of Job Satisfaction among Beginning Librarians," Library Quarterly 49:283-302 (1979) . 37. Smith, Kendall, and Hulin, Measurement of Satisfaction, p. 96. 38. Lynch and Verdin, "Job Satisfaction in Libraries," p .442 . 39. Kavanagh and Halpern, "The Impact of Job Level," p .72. 40. Smith, Roberts, and Hulin, "Ten Year Job Satisfaction Trends." 41. Robert I. Sutton and Anat Rafaeli, "Characteristics of Work Stations as Potential Occupational Stressors," Academy of Management Journal30:270 (June 1987). 42. Michael J. Smith, Barbara G. F. Cohen, Lambert W. Stammerjohn, Jr ., and Alan Happ, "An In- vestigation of Health Complaints and Job Stress in Video Display Operations," Human Factors 23:387-400 (1981). 43 . Martell, "Quality of Working Life," p.441.