Attitudes About Hard Work: A Global Perspective on the Beliefs of Government Employees

ABSTRACT

Government workers are often criticized for holding views that are at odds with the mainstream. Few studies have empirically tested the congruency between attitudes held by government and private sector workers. Using data from the 2009 International Social Survey Programme that includes individual responses across 32 countries, we examine whether government employees’ beliefs about opportunities in society systematically differ from those of private sector workers. We estimate the effect of sector (government/private) and position (supervisor/non-supervisor), among workers in “helping professions,” on perceptions of the role of hard work for getting ahead in society. We find that government supervisors and government/private non-supervisors are significantly less likely than private sector supervisors to emphasize the role of hard work for getting ahead. Private sector supervisors believe that hard work is essential for getting ahead. Government workers are less likely to emphasize the importance of hard work and are more inclined to look beyond the individual for explanations of societal success and failure.

Notes

Notes: Sample includes workers ages 25–61 years old. All results are weighted by country and demographic-specific survey weights, when available.

Data: International Social Survey Programme 2009 (ISSP Research Group 2012).

Notes: N = 4,459. Sample includes workers ages 25–61 years old. All results are weights by country and demographic-specific survey weights, when available.

Notes: Sample includes workers ages 25–61 years old. All results are weighted by country and demographic-specific survey weights, when available. Standard errors in parentheses and clustered by country.

Data: International Social Survey Programme 2009 (ISSP Research Group2012).

Significance levels,

^** p < 0.01;.

^* p < 0.05;.

⁺ p < 0.1.

Notes: Sample includes workers ages 25–61 years old. All results are weighted by country and demographic-specific survey weights, when available. Standard errors in parentheses and clustered by country.

Data: International Social Survey Programme 2009 (ISSP Research Group 2012).

Significance levels,.

^** p < 0.01;.

^* p < 0.05;.

⁺ p < 0.1.

Notes: Sample includes workers ages 25–61 years old. All results are weighted by country and demographic-specific survey weights, when available. Standard errors in parentheses and clustered by country.

Data: International Social Survey Programme 2009 (ISSP Research Group 2012).Significance levels,.

^** p < 0.01;.

^* p < 0.05;.

⁺ p < 0.1.

Notes: Sample includes workers ages 25–61 years old. The means are weighted by country and demographic-specific survey weights, when available.

Note: ISCO-88 is the International Standard Classification of Occupations, developed by the International Labour Organization.

NOTES

Two fields of study in public administration have examined the degree of congruence between the attitudes and motivations of government employees and the public. The theory of representative bureaucracy rests on the assumption that there is congruence between the attitudes of minority and female government workers and minority or female clients, which can lead to the employees advocating for the clients. Several studies have confirmed this congruence (Dolan 2000; 2002; Meier 1975; Meier and Nigro 1976; Bradbury and Kellough 2008; Salzstein 1979). Additionally, there are numerous studies that highlight possible differences in motivations between private sector and public sector employees (Rainey, Backoff, and Levine 1976; Perry and Wise 1990; Perry 1996; Vandenabeele, Hondeghem, and Steen 2004; Lewis and Frank 2002).

Details on the ISSP can be found at http://www.gesis.org/en/issp/issp-home/.

Estonia was dropped from the analysis, as they did not ask whether the respondent held a supervisory role. The Czech Republic was also dropped from our sample since they did not classify any respondents as government workers. The majority of the other reductions in the sample are missing responses for the same demographic question. We also reduce the sample by limiting our analysis to individuals whose age is between 25 and 61 years old.

Our matched sample limits occupations to only those who are adequately represented in each sector; splitting the sample in this way still allows us to estimate unbiased parameter effects on the role of hard work on sector and position while controlling for occupation. We evaluate the attitudes of workers who are reasonably likely to work in either sector, thereby making the analysis more robust. For example, there is no reason to suspect that a heterogeneous group of private sector workers with no counterpart in government should serve as a valid point of comparison for attitudes about the importance of hard work. Additionally, our matched sample recognizes that there are many different types of government occupations. Workers in the occupations we examine interact with clients and therefore we expect their sector of employment or their work experiences to influence their attitudes on hard work.

Concerns that responses to the hard work measure by U.S. respondents drive our results are unfounded. Separate analysis focusing exclusively on the U.S. does indicate a strong effect of both public sector employment and supervisory employment in the U.S. However, our main findings hold even when U.S. respondents are removed from the sample. Additionally, in our preferred specification we include country-level dummy variables as controls.

Supervisors answer yes to “Does the respondent supervise others at work.” A small number of cases (272 out of 4,459; approximately 6%) were missing responses to the supervisor question. We used multiple imputation to assign missing values rather than drop those observations from the sample. Estimates that drop observations due to missing values are nearly identical in terms of parameter estimates and levels of statistical significance.

In most countries, weekly hours are reported as the “usual hours” worked on the primary job. In other countries, hours worked are reported as “actual hours” worked in the previous week on all jobs. Given the differences in how this question is asked, the estimated variance in hours worked will differ by country; to cope with this, we cluster the standard errors. In 13.9% of cases we impute work hours for observations that are missing this measure.

We test other measures of social and economic inequality, such as an individual's perception of inequality, differences between the desired level of inequality and actual levels, and social position in society; none of these variables were related to hard work, nor did they alter our findings.

Views about the role of hard work in getting ahead might predict weekly work hours; if this is the case, then endogeneity would provide biased estimates. It may be that people who believe hard work is the most important predictor of getting ahead choose to work additional hours. To address this concern, we also estimated the models excluding weekly work hours; our results remain virtually the same as those reported here. In addition, we do not find a significant difference in our estimates for public supervisors and non-supervisors. However, we recognize that this is a limitation of our analyses.

Note that the marginal effects are based on estimates from the 5 category model: not important at all (1), not very important (2), fairly important (3), very important (4), essential (5). For simplicity, we provide the marginal effects for the last two categories. Marginal effects are average marginal effects (AME), not marginal effects at the mean (MEM).

It is important to remember that our sample consists of individuals working in the “helping” professions, so all of our respondents have regular client interaction.

Government supervisors comprise 9.6% and 13.3% of the older sample. Supervisors in the private sector make up 6.5% and 6.4% of the younger and older sample, respectively.

Additional information

Notes on contributors

JUSTIN B. BULLOCK

Justin B. Bullock ([email protected]) is a doctoral candidate in the Department of Public Administration and Policy at the School of Public and International Affairs at the University of Georgia. He recently accepted the position of Assistant Professor at the Bush School of Government and Public Service at Texas A&M University. His research interests include public management, public policy analysis, bureaucratic discretion, decision-making theory, and research methodology.

JEFFREY B. WENGER

Jeffrey B. Wenger ([email protected]) is Associate Professor in the Department of Public Administration and Policy at the School of Public and International Affairs at the University of Georgia. He earned his Ph.D. in public policy at the University of North Carolina, Chapel Hill. His research areas include public policy analysis, contingent and nonstandard employment, unemployment insurance, health insurance, state employment policy, and moral hazard.

VICKY M. WILKINS

Vicky M. Wilkins ([email protected]) is Associate Professor in the Department of Public Administration and Policy at the School of Public and International Affairs at the University of Georgia. She earned her Ph.D. in political science at the University of Missouri. Her research interests include representative bureaucracy; gender, representation, and institutions; bureaucratic discretion; human resource management; and political institutions.