Remote work and team productivity

Last registered on December 06, 2023

Pre-Trial

Trial Information

General Information

Title
Remote work and team productivity
RCT ID
AEARCTR-0012002
Initial registration date
November 21, 2023

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
December 06, 2023, 7:42 AM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
Ohio University

Other Primary Investigator(s)

PI Affiliation
University of North Carolina Charlotte

Additional Trial Information

Status
On going
Start date
2016-11-06
End date
2024-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Remote work policies remain controversial mainly because of productivity concerns. The existing literature highlights how the remote setting affects individual productivity yet little is known about how the remote setting affects work in teams - where productivity losses are potentially higher given the additional role of beliefs over partner productivity. Our study closes this gap by examining the effort of individuals randomly assigned to work in either a remote or office setting with partners who are remote and office based.
External Link(s)

Registration Citation

Citation
Dutcher, Glenn and Krista Saral. 2023. "Remote work and team productivity." AEA RCT Registry. December 06. https://doi.org/10.1257/rct.12002-1.0
Experimental Details

Interventions

Intervention(s)
Participants are randomly assigned a location (in the lab or working remotely) and teammates. They always engage in the same real effort production tasks.
Intervention (Hidden)
In the experiment, participants choose if they want their output to count towards an individual payment or team payment. The individual payment scheme is a fixed piece-rate in all sessions while the team incentive differs. In some experimental sessions, the team payment is based on a minimum effort payment. In other sessions, the team payment is based on a revenue sharing payment scheme. Because participants were randomly assigned to a session, they were also randomly assigned to a team payment scheme.
Intervention Start Date
2016-11-06
Intervention End Date
2024-12-31

Primary Outcomes

Primary Outcomes (end points)
We are interested in individual and team productivity based on the location of the individual and their teammates as well as the team incentive scheme. One primary outcome metric will be the difference in team contributions (output) when a participant is paired with office-based vs. remote-based teammates. We will test for differences using t-tests and standard (random effects) panel regression analysis. We will also examine overall productivity (individual output + team output) by location of the worker, using t-tests and regressional analysis to determine differences in overall productivity. Last, we will also examine the proportion shared (constructed variable explained below) using t-tests and regression analysis.
Primary Outcomes (explanation)
Proportion shared is constructed by taking the ratio of the contribution made to the team to total output (individual + team). This is a percentage measure for how much is shared with the team out of all output.

Secondary Outcomes

Secondary Outcomes (end points)
The minimum effort pay scheme game has a prediction that the amount given to the team directly measures beliefs of partner productivity. We also measure beliefs using a survey question at the end of the experiment as a comparison to this explicit measure. The revenue-sharing game does not have a belief measurement directly taken from the game, so we only have the survey question at the end to measure. In all cases, we ask whether the remote or office worker partners contributed more to the team (a direct binary measure of productivity). In the minimum effort treatment, we also ask an incentivized belief question about the minimum amount contributed to the group account by both types of partners. We ask the same question in the revenue sharing treatment but shift from the minimum to a question about the average contribution of teammates. We use all of these questions to examine stated beliefs and to use these beliefs to explain work behavior in the team. We use these beliefs to isolate different types of subjects: those who believe office workers are more productive, those that believe remote workers are more productive and those who believe there are no differences in productivity by work location. We also measure ability in a single round at the start of the experiment, where subjects are paid a piece rate for completed tasks. This is used as an explanatory variable in regression analysis.

Earlier version of this experiment were not registered and they were run prior to Covid. A secondary outcome of interest is whether or not there is a change due to covid. Only the minimum effort treatments were conducted before Covid. Revenue sharing treatments and additional minimum effort treatments will be conducted post-pandemic.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Participants are asked to complete a real effort typing task. The work is performed on an online platform which guides all participants through the experiment. The experiment is self-guided for both office workers and remote workers to ensure that the work platform and information given to all participants is identical. After a set of instructions and an unpaid practice work exercise, the paid portion of the experiment begins with an individual work round, followed by two team work rounds. The experiment ends with a basic survey.
Experimental Design Details
There are two types of subjects in our experiment: remote workers (R) and office workers (O). Remote workers are subjects who participate online in a location of their choice any time in a block of 24 hours. Office workers are subjects who also participate online, but at a pre-specified time and location in a university computer laboratory. In the experiment we use the following phrasing to introduce types: There are two types of participants in this experiment: 1. Those that participate in a computer lab at Ohio University at a designated time (LAB PARTICIPANTS) 2. Those that participate in a location of their choice (e.g. in their apartment, at a coffee shop, in the library, etc.) at any time within a block of 24 hours (TELECOMMUTERS).

Participants are asked to complete a real effort typing task. Each task requires typing a 20-character string of letters and numbers. Subjects are not allowed to continue to the next string until the current string is typed correctly. The same sequence of random strings is used across rounds. The work is performed on an online platform which guides all participants through the experiment. The experiment is self-guided for both office workers and remote workers to ensure that the work platform and information given to all participants is identical. After a set of instructions and an unpaid practice work exercise, the paid portion of the experiment begins with an individual work round, followed by two team work rounds. The experiment ends with a basic survey.

The individual work round lasts for 5 minutes. Decisions are framed as submitting completed tasks to an individual account, which pays $0.10 per correct task. As an outside leisure option, subjects are instructed that they can browse the web, or engage in any other activity besides the typing task. Given their flexibility of participation, remote workers have a wider range of alternative activities than office workers.

Following the individual round, all subjects enter into the two team work rounds, each lasting 20 minutes. In each team round, a subject is randomly paired with two others for a team of three. All members in a team perform work individually using the typing task used in the individual round. The payment scheme in the team rounds includes the individual account seen in the first round that continues to pay $0.10 per correct task, and also includes a team account. In the minimum effort payment sessions, the team account pays $0.18 to the minimum number of correct tasks contributed by a team member. In the revenue sharing sessions, each correct task completed for the team pays $0.18 to the team, and this is equally divided between the team members creating a social dilemma common in these pay schemes. A subject chooses, in real time, to contribute the completed task to the individual account or the team account. Subjects are again instructed that they can browse the web, or engage in any other activity besides the typing task, and they are informed that their partners have this same option.

Our main treatment is the location type of partner assignment in team rounds. In one team round, subjects, who are either O or R type, are paired with two remote workers (RR) and in the other team round, they are paired with two office workers (OO). To control for order effects, in some sessions subjects are paired with two remote workers first and in the remaining sessions subjects are paired with two office workers first. The locational composition of every team member in both rounds is announced prior to the beginning of the team work rounds so that subjects know the location composition of the team they are participating with in the upcoming rounds prior to the start of the first team round. During the team rounds, they are also given a reminder about the location of their current team members at the top of their screen. All other feedback is restricted, including team payoffs. To accomplish the matching scheme where the subjects of interest are partnered in team rounds with two office workers (OO) and then two remote workers (RR), we recruited an additional pool of subjects to serve as partners. Partner subjects include both office worker and remote worker types and the basic structure of the experiment is identical to the primary subjects other than the location of teammates.
Randomization Method
We will use random.org for determining the initial sessions to conduct. We aim for a balanced sample and as such the final sessions may be run on an as-needed basis. Subjects do not know which treatment they are in when they sign up to take part in the study.
Randomization Unit
Individual
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
0
Sample size: planned number of observations
310
Sample size (or number of clusters) by treatment arms
165 in each team payment scheme.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Ohio University IRB
IRB Approval Date
2016-05-09
IRB Approval Number
16-X-137
Analysis Plan

Analysis Plan Documents

Additional details PAP

MD5: 92274b64ab49d8d1eccfe39edbc68219

SHA1: 342af35444ed7a0f2f11a8abcbad48948e1c79fe

Uploaded At: November 21, 2023

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials