The Effect of Telework and Remote Work on Key FEVS Outcomes

Date published: January 16, 2025

Why did we do this study?

Previous studies using the OPM Federal Employee Viewpoint Survey (FEVS) have shown positive effects of telework and remote work on many aspects of organizational climate and employee satisfaction. However, those effects may be due to inherent demographic and professional differences between teleworkers, remote workers, and onsite workers.

This study assessed the effect of telework and remote work on FEVS outcomes using a matching approach to produce greater causal evidence on how remote work and telework shape perceptions of organizational climate and employee engagement. Using the 2023 FEVS, we compared key outcomes between employees who share agency, employment, and personal demographics but differ on their remote or telework status.

What approach did we use?

We used exact matching on FEVS categories to adjust for inherent differences between employees who engage in remote and/or telework and onsite employees. Survey respondents were matched on their agency (and subcomponent for large and very large agencies), pay category or grade level, supervisory status, age category, gender, race/national origin, disability status, length of Federal tenure, and if their occupational family is considered to be either trade, craft, or labor or white collar. The exact matching relied on demographic categories included in the FEVS, which are already grouped.

Employees who matched on their personal and professional demographics but differed on their level of remote work, telework, and onsite work were the basis of comparison. The study used four distinct matched samples to describe how outcomes differed by employee schedule. Table 1 details the four comparisons made throughout the study.

Table 1: Definitions of comparisons constructed based on responses to FEVS question, “Please select the response that BEST described your current remote work or teleworking schedule.”

Comparison	Group with More Frequent Telework or Remote Work	Reference Group
Fully remote vs. fully onsite	Approved remote agreement	No telework (for any reason)
Majority telework vs. fully onsite	3+ days a week of telework or remote work	No telework (for any reason)
At least weekly telework vs. no or infrequent telework	1+ days a week of telework or remote work	Less than weekly telework or no telework
Fully remote vs. regular telework	Approved remote agreement	Telework 1 or more days per week (excluding fully remote)

For differences in average index scores, we used g-computation to estimate the average treatment effect for each matched sample. For single items, we used logistic regression to calculate the odds ratio for a positive response (for example, ‘Agree’ or ‘Strongly Agree’ on the 5-point Likert scales used by FEVS) for each item and for each matched sample. All reported differences are statistically significant at p < 0.05.

What did we learn?

Differences in average FEVS index scores

Remote employees and employees who telework more frequently consistently report higher average scores on FEVS indices compared to similar employees who work onsite more frequently. The difference in average FEVS index scores were most pronounced when comparing fully remote employees to fully onsite employees. When comparing fully remote employees to employees who regularly telework, we identified no meaningful differences in average index scores.

Figure 1 presents the differences in average Employee Engagement Index (EEI) scores, where fully remote employees averaged 4.1 (out of 5) compared to fully onsite employees averaging 3.7 (out of 5). When comparing fully remote employees to employees who regularly telework, there is only a small difference of 0.1 between the two groups.

Figure 1: Predicted average composite EEI scores by level of telework.

Figure 1 Description

The results for other FEVS indices parallel those of the EEI, though the total difference in estimated average index scores varies by index. For the Employee Experience Index (EXI), fully remote employees averaged 4.1 (out of 5) whereas fully onsite employees averaged 3.8 (out of 5). For the Global Satisfaction Index (GSI), fully remote employees averaged 3.9 (out of 5) whereas fully onsite employees averaged 3.5 (out of 5). For the Diversity, Equity, Inclusion, and Accessibility (DEIA) Index, fully remote employees averaged 4.1 (out of 5) whereas fully onsite employees averaged 3.7 (out of 5). For the Performance Confidence Index, fully remote employees averaged 4.4 (out of 5) whereas fully onsite employees averaged 4.1 (out of 5). Comparing fully remote employees to employees who regularly telework consistently yielded no meaningful difference (approximately 0.1) for all indices.

Differences on FEVS single items

Similar trends were observed for differences in single item responses, where remote employees and employees who telework more frequently had higher odds of responding positively on many FEVS single items compared to employees who work onsite more frequently, with only modest differences between fully remote workers and majority teleworkers.

Table 2 presents the logistic regression results for select single items for each matched sample. In response to the question, “Employees in my work unit consistently look for ways to improve customer service”, the odds of a fully remote employee responding positively were 2.6 times higher than the odds of fully onsite employees. For the same question, the odds of a fully remote employee responding positively were only 1.3 times higher than the odds of employees who regularly telework. For other single items included in the study, we identified similar trends.

Table 2: Definitions of comparisons constructed based on responses to FEVS question, “Please select the response that BEST described your current remote work or teleworking schedule.”

FEVS Question	Fully Remote vs. Fully Onsite	Majority Telework vs. Fully Onsite	At Least Weekly Telework vs. No or Infrequent Telework	Fully Remote vs. Regular Telework
Q15: The people I work with cooperate to get the job done.	2.4	2.4	1.6	1.3
Q31: Employees in my work unit approach change as an opportunity.	2.2	2.0	1.5	1.4
Q33: Employees in my work unit consistently look for ways to improve customer service.	2.6	2.2	1.6	1.3
Q66: Management involves employees in decisions that affect their work.	2.4	2.0	1.4	1.4
Q69: How satisfied are you with the recognition you receive for doing a good job?	2.3	2.1	1.4	1.3

Note: The gradation of table cells reflects the magnitude of the odds ratio with darker blue representing higher odds ratios and lighter blue representing odds ratios closer to 1.0. An odds ratio of 1.0 would indicate that the odds of responding positively are equal for each group in the comparison.

Differences in intentions to leave

FEVS asks respondents about their intention to leave their current organization within the next year. This question asks respondents intending to leave to indicate if they plan to retire, take a different job in the Federal Government, take a job outside of the Federal Government, or leave for other reasons.

Remote employees and majority teleworkers have modestly lower odds of indicating that they intend to leave their current organization within the next year. Table 3 depicts the logistic regression results for if respondents indicated that they intend to leave their organization and for what reason for each matched sample. When comparing the intention to leave between employees who regularly telework and employees who telework infrequently or not at all, there is little meaningful difference.

Table 3: Calculated odds ratios for turnover intention responses by level of telework.

Reason For Leaving	Fully Remote vs. Fully Onsite	Majority Telework vs. Fully Onsite	At Least Weekly Telework vs. No or Infrequent Telework	Fully Remote vs. Regular Telework
Leaving for any reason, including retirement	0.8	0.7	0.9	0.8
Leaving for any reason, excluding retirement	0.7	0.7	0.9	0.7
Leaving to take another job	0.7	0.8	0.9	0.7
Leaving to take another job within the Federal Government	0.7	0.8	1.0	0.7
Leaving to take another job outside of the Federal Government	0.8	0.6	0.8	0.8

Note: The gradation of table cells reflects the magnitude of the odds ratio with lighter orange representing odds ratios closer to 1.0 and darker orange representing lower odds ratios. An odds ratio of 1.0 would indicate that the odds of reporting the reason for leaving the organization are equal for each group in the comparison.

What does this mean?

Overall, the results from this study suggest that fully remote employees and employees who frequently telework report greater employee satisfaction and more positive perceptions of organizational climate than fully onsite employees or employees who do not telework on a regular basis. Additionally, fully remote employees and employees who frequently telework are less likely to indicate that they intend to leave their organization within the next year. The findings from this study suggest that the availability of workplace flexibilities like remote work and increased telework may be effective in improving Federal employee satisfaction, and in turn, decreasing organizational attrition.

Back to Top