The Innovation Journal: The Public Sector Innovation Journal, 5(1), 2000, article 1.


A Comparative Analysis of Job Satisfaction PDF


Among Public and Private Sector Professionals


David Barrows and Tom Wesson*

Table of Contents

-Downsizing and Restructuring
-Measuring Job Satisfaction
-Table 1: Summary of Studies Comparing Private and Public Sector Job Satisfaction
-Survey Design
-Survey Administration
-Response Rate
-Response Bias
-Sectoral Differences in Responses
-Table 2: Summary of Survey Responses and Tests of Differences in Means
-Correlations Among Responses
-Gender Differences
-Table 3: Gender Differences in Responses


The purpose of this study is to compare the views of public and private sector professionals toward their work in light of the radical changes in the administrative environment that have occurred during the 1990s. In the course of the study, we develop and test a questionnaire to examine how the environment of downsizing and reorganization in the 1990s has affected the productivity, morale, managerial attitudes and practices of senior professionals in the private and public sectors. Before presenting our survey results, however, we will present short overviews of the existing literature regarding both downsizing and restructuring and comparisons of public sector and private sector job satisfaction.


Downsizing and Restructuring

Downsizing and restructuring (D&R) are practices which have become increasingly prevalent in the past decade, affecting most occupational sectors, encompassing all employment positions and impacting on various lines of business. The effects of downsizing have been particularly significant in the public sectors of most OECD countries. By forcing efficiency with fewer resources, fiscal constraints have increased the demand for the delivery of more effective and better quality public programs and services. In many cases, this demand has translated into a smaller, more flexible civil service. The emphasis on managerial accountability, transparency in public spending, and alternative service delivery has resulted in a leaner public sector. So too has the recent focus on policy frameworks, consolidated budgeting processes and the trend towards more flexible pay and staffing requirements in the public sector.

For example, in Ontario, the Ontario Public Service (OPS) has outlined its plans for D&R in its "Framework for Action" (The Government of Ontario, Secretary of the Cabinet, 1997). The key aspects of D&R included in this document are: focussing on core businesses; ensuring quality service to the public; achieving a smaller and more flexible public sector; becoming more integrative and cohesive; and, finally, securing better standards of accountability in terms of how public resources are allocated. To some, logic dictates that downsizing is an effective means of achieving these ends. In an era of consolidation, intense competition and advanced technology, there is pressure to reduce expenses. Over the past five years, these trends have resulted in an OPS which has been reduced by over 15,000 employees (approximately 15 to 20%) and by 12 ministries.

The D&R initiatives undertaken in Ontario are typical of both public and private sector organizations around the world. Many explanations have been offered in an attempt to understand the increased exercise of D&R. Of course, many of these revolve around economic realities such as increasing competitive pressures, the forces of globalization, the effects of new technologies on productivity and so on. However, McKinley, Sanchez and Schick (1995), drawing on institutional theory, suggest that three types of social forces help to explain the prevalence of downsizing in recent years. These forces are constraining forces, cloning forces, and learning forces. Constraining forces are those which pressure decision-makers to do what appears to be the "right thing" and to keep abreast with contemporary organizational trends which, in today's terms, would involve becoming more efficient and effective. Cloning forces are the result of imitating the steps taken by other organizations. This imitation is often labelled "benchmarking". Finally, learning forces are generated when an organization interacts with the network of professional organizations, academic institutions and consultants in its environment and "learns" what are currently viewed as best practices.

Whether the reasons for the spread of downsizing are economic or social, theoretical or practical, however, the fact remains that this practice appears to be becoming permanent. At the beginning of the decade, it was unusual for a white collar worker to experience job displacement as a result of D&R. Today, especially in the public sector where positions were once revered for their apparent security, job cuts and lay-offs are a matter of course. This drastic change in job security begs an examination of the impact of D&R and its implications for productivity, job satisfaction and the well-being of employees. The question then becomes, how has the practice of D&R impacted on the morale of those affected?

A 1994 American Management Association Report indicated that in the private sector decreased morale is one of the most probable effects of D&R. Decreased morale occurred in 86% of the companies studied by the Association (Allen et al., 1996). Another study by Bennett (1991) found that two thirds of private sector firms reported that morale was seriously affected by downsizing.

Layoff Survivor Sickness is a term coined to describe the guilt, lack of organizational commitment, and fear that survivors often experience following a reorganization (Noer, 1993). Allen et al. (1996) set out to prove the hypothesis that survivor attitudes would be negative over a period of time but that these individuals would eventually adjust to organizational restructuring. In doing so, not only did he find that time did impact on employee attitudes to restructuring, it was also found that the career stage of the employee influenced the impact that restructuring had on that individual. Specifically, it was found that managers early on in their careers were more negatively impacted than mid-career managers. Allen et al. offered three possible explanations for this trend. The first is that early career managers have higher expectations and therefore experience higher levels of disappointment when these expectations are not met. Secondly, mid-career managers may have more experience with the restructuring process and may therefore feel more comfortable with organizational changes. Finally, as mid-career managers are generally in higher positions in an organization, they may have greater access to information giving them a better understanding of the change process.

While almost all of the existing research on downsizing has focussed on private sector firms, as was discussed above, downsizing is also very prevalent in the public sector. The effects of public sector downsizing are of particular interest for several reasons. First, the public sector is highly labour intensive and relies upon professionals from all fields. In fact, government is one of the largest employers of professional workers (Emmert and Taher, 1992). Furthermore, the public sector relies on the expertise of such professionals to effectively implement government policies. No matter how well-designed a policy or piece of legislation may be, the results of such policies are contingent on how well they are implemented. The recruitment and retention of qualified employees, then, is an integral part of the operations of any civil service. It has often been posited that the low pay, limited flexibility and limited opportunities for promotion are characteristics of the public sector which prevent the most qualified workers from remaining in government agencies and rising to the top. The result can be a loss in productivity and initiative in the public sector. These hindering characteristics also discourage young people from engaging in civil service careers. A recent Executive Institute Alumni Association Survey in the United States indicated that public careers are not recommended for young people (quoted in Emmert and Taher, 1992).

The downsizing of the public sector also carries with it potential economic costs. As more and more government employees are displaced and turnover rates increase, the costs of retraining and repositioning employees grows steadily. There is also concern that the effects of the reduced morale which often accompanies D&R could have grave consequences in the public sector. The popular caricature of the bloated and bureaucratic civil service greatly belittles the essential role civil servants play in both policy design and implementation in modern democracies. If our civil service is left disgruntled and discouraged by poorly conceived and executed D&R initiatives, the effectiveness of our government will be greatly reduced.

In addition to the effects mentioned above, potential increases in employee absenteeism stemming from downsizing and restructuring can create further costs in productivity and expense. The 1993 World Labour Report by the UN's International Labour Organization claimed that stress has become one of the most serious health issues of the 20th century. The report also points to the staggering costs that result from job stress. One study indicated that stress related diseases such as ulcers, high blood pressure and heart attacks cost the US economy $200 billion a year in compensation claims and medical expenses.

The potential economic and social costs of public sector downsizing are clearly sufficient to warrant a closer examination of the current trend towards creating a leaner public service. Despite the importance of this topic and the role of the civil service in the formulation of public policy, there is relatively little information about public sector professionals and their attitudes toward their work, their managers or their employers. Furthermore, the literature that exists is varied and inconclusive. One issue on which there is apparent agreement is the lack of data available on this topic. There is a barely palpable literature comparing the attitudes of employees in each of the public and private sectors (Weaver and Franz, 1992). Although empirical studies concerning this subject are increasing in numbers, there remain large gaps in the literature.

Measuring Job Satisfaction

In order to measure attitudes of public sector managers - either in isolation or in comparison to private sector managers - it is necessary to adopt an instrument designed for this task. There have been a number of analytical tools developed in an attempt to capture job attitudes and/or satisfaction levels. Emmert and Taher (1992) use Hackman and Oldham's Job Characteristics Model to evaluate job satisfaction. This measure considers the following job elements; skill variety, task identity, task significance, autonomy, and feedback on job performance. The Job Descriptive Index (JDI; Smith, Kendall and Hulin, 1969) is another analytical tool which is often regarded as the most well-developed instrument for measuring job satisfaction levels. The JDI consists of sub-scales for pay, promotion, people, supervision, and work. These tools, and many others like them, have been used in a variety of studies aimed at capturing various dynamics of job satisfaction.

Although there is clearly no closure on the issue, several studies have focussed on determining whether or not there is a meaningful difference between the public and private sectors in terms of job satisfaction. Before examining these studies, it must be noted that there is some contention regarding whether or not the public-private distinction is a valid or useful one. Perry and Rainey (1988) critique this distinction on the basis that there are a variety of definitions used to describe the differences between the public and private sectors, that there are multiple uses to which the distinction is applied, and that the state of the organizational and managerial research related to the distinction is segregated. Nonetheless, empirical studies exist which suggest that the public-private distinction has some credibility. Selected studies of these distinctions are summarized in Table 1.

Table 1: Summary of Studies Comparing Private and Public Sector Job Satisfaction


Methodology Findings
Paine, Carroll & Lecte (1966)




Public managers were much less satisfied with their jobs than similar groups in the commercial and industrial sectors.
Rainey (1979) Job Descriptive Index Private sector managers scored higher in job satisfaction than their public sector counterparts.
Solomon (1987) Two questionnaires were administered to 240 top managers from a variety of private and public organizations. Private sector managers reported consistently higher levels of job satisfaction than public sector managers.
Kumar & Acharnamba (1993) Three hundred employees from public and private industries (workers and administrative staff) completed the Job Descriptive Index. Public sector administrative staff have greater job satisfaction than private sector administrative staff.
Scheider & Vaught (1993) Job descriptive index, completed by 36 private sector employees and by 68 public sector employees. No significant difference in job satisfaction levels between public and private sector employees.

It is apparent that the results concerning the impact of employment sector on job satisfaction are inconclusive. Looking across studies we can see some evidence that public sector job satisfaction has improved relative to the private sector over the last twenty to thirty years. However, for several reasons this conclusion must be made with extreme caution. First, there are many other factors which impact on job satisfaction levels. Second, this literature is predominately based upon US studies. Finally, the studies use a variety of methods and subject populations.

Among the factors which influence job satisfaction is the respondent's level in the organizations. The literature regarding the impact of job level on satisfaction is decidedly mixed, making it difficult to draw any strong conclusions. The most appropriate summary to be made may be that job status is a not a credible determinant of job satisfaction. There is evidence, however, which suggests that job status may be a mitigating factor in determining the degree to which an individual employee is affected by organizational restructuring.

There is also evidence to suggest that the impact of restructuring is also mitigated by other factors such as union and organizational support. Greenglass and Burke (1998a, 1998b, 1998c), in three research papers examining D&R in Ontario hospitals and its psychological impacts on nurses, note several factors that can reduce the perception of negative outcomes. Nurses at affected hospitals reported fewer negative feelings related to D&R when they had a high degree of support from their union throughout the D&R process. This was most evident in situations where restructuring was most significant and the threat to personal employment was high.

Similarly, organizational support appears to lessen the perception of negative outcomes for those experiencing restructuring and downsizing. Greenglass and Burke argue that companionship and open, honest communication is key to forming positive global beliefs amongst employees about an organization. The converse of this is also true, they state, in that employees are more likely to perceive threats to themselves where there is little support for what they do or what they need in order to provide good service. In summary, Greenglass and Burke state the following factors as key to ensuring good morale during restructuring and downsizing initiatives:

1. A vision of what the organization will look like post-restructuring
2. Complete, timely and repetitive communication with employees
3. Recognition of the organization’s past accomplishments
4. Celebrating the organization’s current accomplishments
5. Using staff reductions only as a last resort
6. Investing in training and development of staff, and
7. Career transition assistance to facilitate employment elsewhere.

Greenglass and Burke appear to confirm other researchers’ findings that the highest degrees of anger and hostility toward restructuring and downsizing are found amongst middle managers; particularly where the level of service provided to clients appears to be declining. Furthermore, anger levels seem to be related to tenure and hierarchical level in these circumstances. Contributory factors include the clarity of the vision for the organization, whether the organizational leadership is demonstrably committed to the stated goals, and whether the new direction is respectful of the organization’s past accomplishments.


Increasing levels of anger are one of the symptoms of burnout. According to the International Labour Organization (1993) burnout is becoming more and more prevalent in workforces around the world, affecting British miners and Swedish waitresses just as it impacts Japanese school teachers and U. S. executives.

A study by Golembiewsk et al. (1998) utilized survey data from 34 sites to estimate how many people have varying degrees of burnout in public sector organizations world wide. The survey instrument used in this study was based on an eight phase model of burnout. The model does not contend that individuals progress linearly through the phases, it merely describes them in a range from least severe (Phase I) to most severe (Phase VIII).

Golembiewsk et al. found that there exists a common path through the phases that is frequently followed by employees experiencing burnout. A high depersonalization phase (phase II) deprives individuals of important information, which over time can impede performance on task (phase IV) and these two conditions can coexist at levels beyond the individuals’ coping abilities, leading to emotional exhaustion (phase VIII). The major findings of this research were threefold: a large number of public servants are in advanced stages of burnout; North American public sector employees experience levels of burnout similar to those in the private sector; and the distribution of phases is bipolar.

In another view of burnout, Maslach (1993) has identified three major elements or dimensions of burn-out: a feeling of emotional exhaustion; a feeling of depersonalization and detachment from those around onself; and a feeling of reduced personal accomplishment. Maslach views burnout, particularly the depersonalization aspect of it, very much as a social phenomenon, one stemming from and impacting upon the interactions between the person feeling burnt-out and those around him or her. This view stems in part from the fact that Maslach developed her theories of burnout largely through studies of health care professionals. However, clearly public servants in general are subject to job stresses similar to those experienced by health care providers, albeit often less severe. Like health care workers, public servants, particularly public sector professionals, are often working for the benefit of specific individuals or groups (constituencies). As job pressures increase, they may be prone to see members of such constituencies as objects rather than as individuals.


Given the existing literature on D&R, job satisfaction and burnout, the goal of this study is to investigate differences in job satisfaction and job attitudes between public and private sector lawyers in Ontario. The basic research strategy employed in the study was to administer a mail survey of samples selected from the subject populations.

The decision to employ a mail survey was made for a number of reasons. First, this approach was within the study’s budget constraints and allowed us to distribute surveys to a number of different locations around the Toronto and Ottawa areas without encountering any increase in costs. Second, this approach is considered to be most appropriate for dealing with sensitive topics and we felt that the subject matter in our study of managerial attitudes was of a sensitive nature (Kervin, 1992). Third, mailing the surveys allowed us contact with otherwise difficult to reach potential respondents; the "on the move" nature of the work, time constraints of lawyers and employment status of individuals targeted in our study made this an attractive attribute (Emory and Cooper, 1991).

Survey Design

Our initial intention was to develop a survey instrument which closely followed the methodology of Zussman and Jabes’s (1989) landmark study of the federal civil service. However, in our estimation Zussman and Jabes’s survey was far too long for our purposes. We were aware that by not using a survey similar to Zussman and Jabes’ we would lose some of their ability to explore relatively subtle aspects of job satisfaction and attitudes. Nonetheless, our preliminary discussions with potential respondents and our survey pre-tests convinced us that if we were to achieve our desired response rates we would have to have a survey which could easily be completed in about 10 minutes. This necessitated a survey of about 40 to 50 items.

Once we made the decision that we needed a survey significantly shorter, and therefore different, than Zussman and Jabes, we realized that there was little to be gained by modelling our survey after theirs. We could not overlap with their survey enough to allow us to compare specific results with theirs, so we decided to look elsewhere for an appropriate means of measuring job satisfaction. While numerous measures of job satisfaction exist, we elected to employ the Job Satisfaction Survey (JSS; Spector, 1997). The JSS was chosen for several reasons. First, although it is short (36 questions) it yields not only an overall measure of job satisfaction, but measures of satisfaction on 9 sub-scales as well. Additionally, the JSS is freely available for use for academic studies and the author has published norms to allow comparisons between the sample group and the general population.

In addition to the questions from the JSS, we wished to include additional questions on our survey relating to burn-out. The literature on burn-out has identified three major elements of burn-out: a feeling that ones life is dominated by one’s work, a feeling of emotional exhaustion and a feeling of detachment from those around oneself. We included one question relating to each of these elements of burn-out in our survey. We also asked two questions concerning an aspect of job satisfaction closely related to burnout - the ability of respondents to strike the right balance between the demands of their work and personal lives.

We also included two simple questions to explore the respondent’s assessment of both their employer’s current level of effectiveness and how that effectiveness has changed over the last 3 years. We took a similarly straightforward and simple approach in addressing a single question at the respondents perception of their employer’s effectiveness in meeting their needs for training.

In another series of additional questions we asked respondents for their assessments of the leadership provided by senior management in their organization. This is an important element of job satisfaction - especially during times of restructuring - which is not directly captured by the JSS. One can easily argue that many of the keys to successful downsizing reported by Burke and Greenglass and others relate directly to the effectiveness of the top leadership of the downsizing organization.

All of the questions described thus far were in a format in which the respondent is asked to indicate on a six point scale the extent to which they disagree or agree with a particular statement. The scale is presented and scored numerically. Responses can range from 1, indicating strong disagreement with the statement, to 6, indicating strong agreement with the statement. The statements are phrased in such a way that in some cases agreeing with the statement would be a favourable response while in others disagreement would indicate a favourable response. Before analyzing the responses to such questions they were rescaled such that a low response would indicate dissatisfaction with an element of one’s work.

In an additional question we asked respondents to tell us which two elements of their job is most important to them in determining their overall level of job satisfaction. The choices correspond roughly to the nine sub-scales of the JSS, plus organizational leadership. This question is important in providing some background against which to assess the responses to the rest of the questionnaire. For example, if the respondents are very satisfied with the benefits they receive, but this element of their work is not important in determining overall job satisfaction, then the fact that they are satisfied with this aspect of their work is of relatively little importance.

The final question in the main body of the survey asked respondents whether they had ever considered a career in the public [private] sector if they are currently employed in the private [public] sector. This question also included a space for respondents to give open-ended reasons for their answer. This question served the purposes of forcing the respondents to think about the pros and cons of work in each sector.

The survey continued with a series of 10 demographic questions which give information on the respondents personal and employment background (age, marital status, income, title, hours worked per week, etc.). The survey concluded by giving respondents an opportunity to provide any other comments they deem appropriate.


The focus of our research is on comparing job satisfaction and attitudes among managers and professionals in the private and public sectors in Ontario. In order to ensure comparability and to facilitate the administration of the survey we targeted lawyers in each of the two sectors for this study.


The survey was pretested by giving it to several lawyers for review. These pre-test respondents confirmed that the survey questions were, for the most part, clearly phrased and would be easily answerable by members of our sample. However, the pre-test group indicated that the survey in its pre-test form was still too long and somewhat redundant. For this reason we decided to shorten it by eliminating some of the more redundant questions from the JSS. We eliminated three of the four questions relating to pay, three of the four questions relating to benefits, two of the four questions relating to opportunities for promotion. While eliminating these questions meant that we would not be able to compare our sample to the published norms for these sub-scales of the JSS (or for the overall satisfaction score), the pre-tests indicated that the redundancy in the questions would seriously reduce response rates. None of these sub-scales are areas of primary focus in the current study.

Survey Administration

The survey was mailed to 150 members of each of the following groups:

- Crown Attorneys and Assistant Crown Attorneys working in the Toronto, North York, Scarborough, Richmond Hill, Peel or Ottawa Crown Attorney offices

- Lawyers employed in 10 large law firms in the Greater Toronto Area.

In each case, the names and addresses of the members of the sample were selected randomly from directories posted on the internet.All surveys were mailed to members of the sample by regular mail to their office address. Approximately one week after the initially mailing all members of the sample were given a reminder call. If the respondent requested, a new survey was faxed or mailed directly to them. A second round of follow-up phone calls was made approximately two weeks after the initial round to all non-respondents.


Response Rate

The survey method described above has yielded 60 responses (40% response rate) from the private sector lawyers and 38 responses (25.33% response rate) from the public sector lawyers. An unfortunate drawback to the mail survey approach is the typically low response rates associated with this method. These response rates are in the range to be expected from this type of survey. Perhaps what is somewhat surprising is that the response rate from the private sector was higher than that of the public sector.

A number of factors are believed to contribute to mail questionnaire survey response rates (Bruvold and Comer, 1988; Goyder, 1982). We considered the factors identified by the study by Bruvold & Comer based on 464 individual survey cases from 200 separate articles encompassing 55 years of published methodological research in major research journals. We addressed as many of these as possible, including, shortening the length of the survey, attaching a return envelope and conducting two follow-up phone calls, one week and two weeks after the survey was distributed. However, some of the questionnaire characteristics themselves which we could not change may also have had a negative affect on the survey response rate. These characteristics included the type of the data requested and the nature of the data requested. In particular, lower response rates are expected when the type of data is attitudinal and when the nature of the data requested concerns anxiety arousing topics (e.g. embarrassing, threatening, sensitive, potentially offensive or confidential matters) (Bruvold and Comer, 1988; DeLamater, 1982). Anxiety often results from the respondents’ awareness of social norms which define these topics as inappropriate matters for conversation. In addition to these factors, the topic importance to the respondent can often have an impact on response rates and this also was something we could not anticipate.

Response Bias

One of the problems with mail questionnaire surveys is in determining non-response bias which may occur when there is a systematic under-representation of certain types of cases in the achieved sample. Because there is less opportunity to assess the extent of this type of bias it is also more difficult to correct for it.

We spoke with a number of individuals in follow-up calls who indicated that they were not intending to participate in the study. One of the most common reasons given was that they did not have time to complete the survey. Because the survey required only 5 to 10 minutes to complete, we considered "not enough time" a proxy for the saliency of the topic; apparently this topic was not of great importance to these individuals. A few others indicated that they had been advised by those in superior positions in their firm that it was not appropriate for them to complete the survey. Thus, anxiety resulting from these social norms discouraging potential respondents from participating most likely contributed to the low response rate. However, within each sector there were no significant differences in response rates among the firms or Crown Attorney offices involved in the survey.

There were a number of non-contacts resulting from potential participants who could not be reached due to the fact that they moved around to different locations on a frequent basis. As well, potential respondents who were on vacation during the summer months when the study was conducted may have also reduced the response rate.

We were also able to compare distributions of the achieved sample with the target population based on gender. This analysis revealed that the response rates in both sectors were essentially identical for males and females.

Sectoral Differences in Responses

The results of our survey indicate significant differences between public and private sector lawyers in Ontario with respect to both intrinsic and extrinsic satisfaction with their job. Furthermore, our findings indicate that differences in the actions and attitudes of top management in these sectors is directly responsible for a large part of these differences in the attitudes of their employees.

Table 2 summarizes our findings for questions relating to each major category of question. The first nine categories relate to the sub-scales of the JSS . The remaining relate to other questions or groups of questions as described above. For each category of question, Table 2 indicates the mean response on the question(s) relating to that category for each of the public sector, private sector and all respondents. The table then shows the f-statistics and significance levels for a test of the difference in the mean responses between the two sectors.

While the most striking feature of Table 2 is perhaps the huge differences in the responses of the private and public sector lawyers, there are several similarities worth noting. Perhaps the most important is that the two sectors have nearly identical levels of satisfaction on the JSS sub-scale which measures respondents satisfaction with the nature of the work they do.

Table 2: Summary of Survey Responses and Tests of Differences in Means




Mean Response

Std. Dev. F-Stat Significance
Nature of the Work Itself Public


Pay Public


Benefits Public


Contingent Rewards Public


Opportunity for Promotion Public




Communication Public
16.170 0.000
Co-Workers Public




Operating Conditions Public




Supervision Public




Leadership Public




Training Public


Problems of Balance - Work & Personal Public




Degree of Burnout Public




Current Organizational Effectiveness Public


Change in Organizational Effectiveness Public


We argue that because the employees in the two sectors are equally satisfied with this most fundamental aspect of their job, they should be able to achieve similar levels of overall job satisfaction.

There are two other aspects of their work in which the two groups did not have significantly different responses: their feelings of burn-out and their feelings about their ability to balance their work and personal lives. Given the responses we received to the open ended questions on the survey it is clear that these results would surprise some of the private sector lawyers, who clearly feel that their public sector counterparts do not work as hard as they do themselves. Both groups felt what could be described as a moderate degree of burnout (average score over all respondents of 3.38 on the three burnout related questions). The private sector respondents did report slightly higher feelings of burnout. Both groups reported higher and near identical levels of agreement with statements indicating a lack of ability to find time for themselves and their families (an average score 4.59 across both groups).

While there are these three areas of similarity between the public and private sector responses to our survey, there are vast differences between the responses of the two sectors. Perhaps least surprising of these differences is in satisfaction with compensation. In terms of both pay and benefits the private sector lawyers agreed much more strongly than the public sector lawyers that they were well compensated for their work. However, what may be more surprising is that these differences between the groups extended beyond monetary rewards and tangible benefits to less concrete factors such as opportunities for promotion and intrinsic rewards contingent on performance such as recognition for a job well done and a feeling that the organization appreciates one’s efforts.

Another aspect of their work with which the public sector lawyers were much less satisfied than the private was in terms of the conditions under which they have to perform their work. The public sector respondents felt that they were bogged down by paperwork and red tape to a far greater degree than the private sector respondents did. Perhaps this finding would be expected by many observers. However, it is interesting that in performing the same type of job one group can feel much more constrained by their organization. Clearly, these constraints are imposed by the organization, not by the nature of the work.

It is also important to reexamine the finding that lawyers in the two sectors had similar attitudes about the nature of their work despite these very different perceptions about the context in which their work was performed. The fact that a perceived oppressive work environment did not cause the public sector lawyers to develop negative attitudes about their work itself speaks volumes about their commitment to and enthusiasm for their profession. Again, we conclude that public sector lawyers should be much more satisfied with their job than they are.

The public sector respondents were also significantly less satisfied with their co-workers than were the private sector lawyers. However, it is important to note that both sectors indicated relatively high levels of satisfaction with their co-workers and that the difference between the sectors, although highly significant statistically, is relatively small in absolute terms.

Several of our sub-scales relate to the respondents attitudes about their superiors in the organization in which they are employed. The communications sub-scale from the JSS deals with such issues as the degree to which the respondents understand the goals of the organization in which they work, the extent to which their work assignments are fully explained to them as well as their general feelings about communications in the organization and the extent to which they feel that they "know what’s going on" in the organization. Clearly, creating an organization in which communications are open and effective is a responsibility of the organization’s managers, with the tone of communications often being set very high up in the organization.

Two more sub-scales deal more directly with management issues. First, the supervision sub-scale of the JSS deals with the respondents relationship with their direct superiors. This sub-scale measures such things as the respondents’ perception that their organizational superiors are fair to them, take their feelings into account in decision making as well as the respondents’ feelings about supervisors’ general competence and the extent to which the respondents like their superiors.

Second, the leadership sub-scale attempts to measure the respondents feelings about their superiors’ performance in terms of the key top management activity of providing leadership. Recall that Burke and Greenglass have shown that several key determinants of the effect of downsizing and restructuring on organizational morale relate to the organization’s leadership.

The survey results show that for all three of these sub-scales the public sector respondents are significantly less satisfied than their private sector colleagues. This is despite the fact that for the private sector respondents both the communications and leadership sub-scales were among those with the lowest overall levels of satisfaction.

Another major area of difference between the two groups of respondents is in terms of their feelings that they are receiving adequate training. For the public sector respondents this question received the lowest average response of any on the survey, indicating that there are vast differences between the training these lawyers think they need and that provided by the government. The private sector attorneys are relatively well satisfied with the training they receive.

A final pair of questions relate to the respondents’ perceptions of their organizations’ overall effectiveness. Here we see not only that the public sector respondents agree to a much less extent than the private sector respondents do with the statement that their organization is effective in achieving its goals but that they are also less likely to agree with the statement that the organization is performing better today than it did three years ago. In fact, this statement is one with which the public sector respondents on average disagree, indicating that the Crown Attorneys responding to our survey see their organization’s performance worsening of late.

Amongst these overwhelmingly negative attitudes expressed by the public sector respondents to our survey there is one other small glimmer of hope. When asked to identify the two aspects of their work which contributed most to their overall level of job satisfaction from among 10 choices both groups of respondents overwhelming selected the nature of the work itself as the most important. In each group more than half of the respondents ranked this aspect as the most important. This is the one aspect of job satisfaction where the public sector respondents were as happy as the private sector respondents. It is also interesting to note that the distributions of responses to this question were very similar for the two sectors. Unfortunately for the politicians who manage our public sector pay and benefits were important aspects of job satisfaction for many of the respondents.

To sum up, our overall findings are indicative of systematic unhappiness with their work environment on the part of the public sector respondents to our survey, particularly in comparison to the private sector respondents. However, respondents from the two sectors are equally satisfied with the actual work that they perform. That is, the differences we see are not differences in satisfaction with job content, but rather a differences in satisfaction with the environment in which the respondents’ work is performed.

Correlations Among Responses

We also performed an analysis of the correlations among the sub-scales for each sector and between the various sub-scales and some of our demographic variables. It should be carefully noted that this analysis cannot address issues of causality. It only looks at the extent to which the two variables tend to move together.

Looking first at the public sector, we found that in general there was a high degree of correlation among the various sub-scales. That is, workers tended to be either satisfied or dissatisfied more or less across the board. The sub-scale with the strongest correlations with other sub-scales was supervision. It had correlations of greater than 0.5 (all greater than 99.9% significant) with all of the other JSS sub-scales.

Among the public sector respondents, all of the JSS sub-scales were also significantly correlated with the respondents’ perception of both their organization’s current effectiveness and its improvement in effectiveness. The weakest correlations were with the pay sub-scale (0.340 with current effectiveness and 0.343 with improvement in effectiveness) and the strongest were with the supervision sub-scale (0.835 with current effectiveness and 0.818 with improvement in effectiveness),the leadership sub-scale (0.787 with current effectiveness and 0.858 with improvement in effectiveness), and the communications sub-scale (0.795 with current effectiveness and 0.752 with improvement in effectiveness). These findings - and the fact that each of these three leadership related sub-scales displays very strong correlations with all of the other sub-scales we examined - are consistent with the view that the leadership and leadership related activities are key factors in determining overall levels of employee satisfaction.

Also of interest in the public sector were the fact that our burnout measure had significant correlations only with our measures of satisfaction with co-workers and with the work itself (-0.526 and -0.546 respectively) and that current annual personal income was not positively correlated any measure of job satisfaction (even controlling for job tenure).

An analysis of the correlations among the private sector scales is also quite interesting. Again, there was a high degree of correlation among the various scales, however the correlations in general were not quite as strong as in the case of the public sector respondents. There is also not as clear a pattern of one or a few scales having greater correlations with the others as there is in the public sector case. Again however, leadership, communications and supervision are among the sub-scales with the highest correlations with respondents perceptions of both their employer’s current overall effectiveness and the change in their employers overall effectiveness over the last three years (correlations with current effectiveness and change in effectiveness, respectively, of 0.352 and 0.545 for supervision, 0.531 and 0.369 for communications and 0.578 and 0.445 for leadership). All of these correlations are highly significant.

In the private sector burnout had a highly significant negative correlation with respondent’s satisfaction with operating conditions within their firm (correlation -0.358) and less significant correlations (less than 99%significance, but still greater than 95% significance) with the respondents’ satisfaction with their supervision (-0.273), the nature of their work (-0.283) and the communications within their firm (-0.339).

Finally, it is interesting to note that in the private sector there were very significant negative correlations between feelings of burnout and both current annual personal income (-0.520) and the respondents years tenure at their firm (-0.502). Since pay and tenure are significantly correlated themselves partial correlations were also calculated. Here we found a partial correlation of -0.4056 (98.7% significance) between burnout and personal income controlling for job tenure but no significant partial correlation between burnout and job tenure controlling for personal income. Thus, it seems that the actual relationship is between income and burnout, not tenure and burnout.

Our major conclusions from this correlational analysis are as follows. First, individual respondents’ satisfaction levels tended to be fairly uniform across the various sub-scales measured. Second, there is support for the hypothesis that the relationship among members of the various levels of the organization - as measured by the leadership, supervision and communication sub-scales - are key determinants of overall satisfaction. Finally, and somewhat surprisingly, feelings of burnout were not associated with across the board negative feelings about one’s work.

Gender Differences

The final set of statistical analyses which we performed looked at gender differences in responses to our survey. First, it is important to note that all of the findings discussed above apply equally to male and female respondents. That is, in the data there are both very strong sectoral differences and strong gender differences in response patterns.

Table 3: summarizes the more important findings regarding gender differences. In the sample as a whole female respondents were significantly less satisfied with their coworkers (difference in means 92.0% significant in a two-tailed t-test if equal variances are not assumed), operating conditions (91.2%), contingent rewards (97.9%), opportunities for promotion (97.3%), and the training opportunities provided by their employer (97.7%). There were also highly significant differences in respondents’ perception of the changes in their organization’s effectiveness over the last three years. In fact, on average the female respondents thought their organization had become less effective over the last three years while the male respondents thought that their organization had become more effective. Finally, the female respondents showed some evidence of having greater feeling of burn-out than their male counterparts (90.2% significance in the difference in the mean responses).

Much more understanding of these gender differences can be gained by looking at them sector by sector. In the private sector the only gender difference that was even somewhat significant was that female respondents felt somewhat more burnt-out than their male counterparts (Mean scale scores of 3.61 and 3.26 respectively. T-test of differences in means 93.4% significant if equal variances are not assumed).

In the public sector, however, female respondents were less satisfied with their co-workers (97.2% significant if equal variances are not assumed), operating conditions at work (90.2%), contingent rewards (97.6%) and opportunities for promotion (97.6%). They also have more difficulties in finding balance between their work and personal lives (96.2%). Finally, they disagree moderately to strongly with the statement that the ministry operates more effectively than it did three years ago (mean response 1.87). Their male counterparts disagree only slightly with the statement (mean response 3.09). The differences in these mean responses is very significant (98.9% significant if equal variances are not assumed).

Table 3: Gender Differences in Responses

Sub-Scale All Respondents Private Sector Public Sector
Contingent Rewards Male Mean
Female Mean
4.13 65
3.30 31
4.68 43
4.65 15
3.05 22
2.04 16
Opportunities for Promotion Male Mean
Female Mean
4.60 63
3.69 28
5.46 41
5.13 15
3.00 22
2.03 13
Coworkers Male Mean
Female Mean
4.87 64
4.52 30
4.98 42
5.07 15
4.68 22
3.98 15
Operating Conditions Male Mean
Female Mean
3.67 64
3.26 31
4.01 42
4.03 15
3.03 22
2.54 16
Leadership Male Mean
Female Mean
3.92 64
3.34 29
4.32 42
4.27 15
3.18 22
2.36 14
Training Male Mean
Female Mean
4.03 64
3.16 31
4.74 42
4.47 15
2.68 22
1.94 16
Problems of Balance - Work & Personal Male Mean
Female Mean
4.52 64
4.81 31
4.65 43
4.47 15
4.26 21
5.12 16
Degree of Burnout Male Mean
Female Mean
3.19 63
3.45 28
3.26 41
3.61 12
3.06 22
3.33 16
Change in Organizational Effectiveness Male Mean
Female Mean
4.14 61
2.67 24
4.74 39
4.00 9
3.09 22
1.87 15

For each sub-scale in each column the three numbers shown on the left are the mean response from male respondents, the mean response from female respondents and the statistical significance of the difference in means. That is, the probability of the sample gender means being at least as different as those shown if the population gender means were in fact equal. In each case, equal variances in responses across genders are not assumed. The two numbers shown on the right are the number of respondents of each gender upon whom the analysis is based.

To summarize, then, in the sample as a whole we have significant differences between male and female respondents. However, these gender differences in the broader sample are driven almost entirely by gender differences in the public sector. This finding is somewhat surprising, given the general feeling that the public sector provides a friendlier work environment for women (and minorities). However, the differences we found may not be the effective poorer response to the demands of female employees in the public sector. Instead, they may be the result of a self-selection effect whereby women with greater awareness of and concern over work-related gender issues elect to work in the public sector. It is still the case, however, that the lower levels of satisfaction displayed by female respondents in our survey must be of serious concern to public sector managers and policy makers.


Restructuring and downsizing are now part of the environment in which professionals function in both the public and private sectors. There is strong evidence in the literature to suggest that downsizing adversely affects morale in the public sector. Associated social and economic costs of decreased morale have also been identified. Burke and Greenglass, among others, offer some insight into how to lessen these effects. However, the literature on this subject is disjointed to say the least.

The results of our survey indicate that while Provincial Crown Attorneys in Ontario enjoy their actual work as much as private sector lawyers, they have significantly lower levels of satisfaction in terms of almost all other aspects of their work life. We find correlational evidence which supports (or at least does not refute) the contention that issues relating to leadership are a major cause for these lower levels of satisfaction. We also found that in the public sector female respondents were strikingly less satisfied than male respondents.

We conclude, therefore, that our study has highlighted several serious causes for concern and has shown the need for further study of the how downsizing and restructuring has been handled in the OPS and what the effects of this process will be on morale and job performance.

We hope in further studies to compare how the restructuring process has been managed in the OPS and various private sector organizations and to undertake a much broader survey of job satisfaction and attitudes of managers and professionals in the OPS and the private sector in Ontario. We also plan, working with partners in the United States and in Germany, to compare how the restructuring process has been managed in different jurisdictions and the effectiveness of these different management styles. This will give much more insight into the optimal way to manage restructuring. While our primary focus is on the public sector, we expect to develop insights which will be of interest to managers in both sectors.

About the Authors:

David Barrows and Tom Wesson are professors with the Schulich School of Business, York University, Toronto, Canada

* The authors would like to thank the York University-Certified General Accountants' Association of Ontario International Business Research Program, the York University Office of Research Administration and the Schulich School of Business for financial support. We would also like to thank John Tibert and Tammy Chi of the Institute for Social Research, York University, and Ian Greene, Janet Cranney and Elizabeth Kurucz for their assistance in conducting this study.

Please send comments and contributions by email to The Innovation Journal


Allen, T.D., D.M. Freeman, R.C. Reizenstein and J.0. Reatz (1996), "Just Another Transition? Examining Survivor's Attitudes Over Time," The University of Tennessee, Department of Management.

Bennett, A. (1991), "Downscoping Doesn't Necessarily Bring an Upswing in Corporate Profitability" The Wall Street Journal, June 4, 1991, pages B-1 and B-4.

Bruvold, N. & J. Comer (1988), "A model for estimating the response rate of a mailed survey," Journal of Business Research, 16 (2): 101-116.

Burke, Ronald J. & Esther R. Greenglass (1998a), "Organization Support during Health Care Downsizing and Restructuring" York University, mimeo.Burke, Ronald J. & Esther R. Greenglass (1998b), "Union support during organizational restructuring," York University, mimeo.

Burke, Ronald J. & Esther R. Greenglass (1998c), "The Relationship Between Hospital Restructuring, Anger and Hostility in Nurses" York University, mimeo.

DeLamater, J. (1982), "Response effects of question content," in W. Dijkstra, and J. Van Der Zouwen (eds.), Response Behaviour in the Survey-interview, Toronto: Academic Press.

Emmert, M.A., W.A. Taher (1992), "Public Sector Professionals: The Effects of Public Sector Jobs on Motivation, Job Satisfaction and Work Involvement' American Review of Public Administration, Vol 22, No. 1.

Emory, C. & D. Cooper (1991), Business Research Methods. Homewood, IL: Irwin.

Golembiewsk, R.T., R.A. Bourdreau, Ben-Chu Sun and Huaping Luo (1998), ‘Estimates of Burnout in Public Agencies Worldwide: How Many Employees Have Which Degrees of Burnout, and with What Consequences?’ Public Administration Review, Jan/Feb, Vol. 58. No. 1.

Government of Ontario, Secretary of Cabinet (1997), Building The Ontario Public Service for the Future: A Framework for Action. Toronto: Queen's Printer for Ontario.

Goyder, W. (1982), "Further evidence on factors affecting response rates to mailed questionnaires," American Sociological Review, 47: 550-553.

Kervin, J. (1992), Methods for Business Research. New York: Harper Collins Publishers Inc.

Kumar, K. Gopi & B. Achamamba (1993), "A Comparative Study of Job Satisfaction and Job Involvement among Public and Private Sector Employees" Psychological Studies, Vol. 38, No. 2.

Maslach, C., (1993), "Burnout: A Multidimensional Perspective," in W.B. Schaufeli, C. Maslach and T. Marek (eds.), Professional Burnout: Recent Developments in Theory and Research. Taylor and Francis: Washington D.C.

McKinley, W., C.M. Sanchez and A.G. Schick (1995), "Organizational Downsizing: Constraining, Cloning and Learning," Academy of Management Executive, 9, 32-42.

Noer, D.M. (1993), Healing the Wounds. San Francisco: Jossey-Bass.

Paine, F.T., S.J. Carroll Jr, and B. Lecte (1966), "Need Satisfaction of Managerial Level Personnel in a Government Agency," Journal of Applied Psychology 50:247-249.

Perry, James L. & Raincy, H.G. (1988), "The Public-Private Distinction in Organizational Theory: A Critique and Research Strategy" Academy of Management Review, Vol. 13, No. 2, 182-201.

Rainey, H. G. (1979), "Perceptions of Incentives in Business and Government: Implications for Civil Service Reform" Public Administration Review, September/October: 440-448.

Schneider, S.D. and R.C. C. Vaught (1993), "A Comparison of Job Satisfaction Between Public and Private Sector Managers" PAQ Spring.

Smith, M.P., L.M. Kendall, and C.L. Hulin (1969), The Measurement of Satisfaction in Work and Retirement: A Strategy for the Study of Attitudes. Chicago: Rand McNally.

UN International Labour Organization (1993), World Labour Report 1993, Geneva: The International Labour Organization.

Solomon, Esther E. (1986), "Private and Public Sector Managers: An Empirical Investigation of Job Characteristics and Organizational Climate" Journal of applied Psychology, Vol. 71, No. 2, 247-259.

Spector, Paul E. (1997), Job Satisfaction: Application, Assessment, Cause and Consequences, Thousand Oaks, California: Sage Publications.

Weaver, Charles N., Franz, Robert S. (1992) "Work-Related Attitudes of Entrepreneurs, Public and Private Employees" Psychological Reports, 70, 387-390.

Zussman, David and Jak Jabes (1989), The Vertical Solitude: Managing in the Public Sector, Halifax: The Institute for Research on Public Policy.

Published (January - April 2000) new pdf (Nov 21 2013)

Last updated: December 28 2018