Worker Voice and Disappointment

Last registered on January 20, 2022

View Trial History

Pre-Trial

Trial Information

General Information

Title

Worker Voice and Disappointment

RCT ID

AEARCTR-0008816

Initial registration date

January 14, 2022

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

January 19, 2022, 12:30 PM EST

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

Last updated

January 20, 2022, 9:34 AM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

United States of America

Region

Primary Investigator

Name

Helen Ho

Affiliation

Harvard University

Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status

In development

Start date

2022-01-17

End date

2022-01-21

Keywords

Labor

Additional Keywords

JEL code(s)

J01

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Firms can harness worker voice to improve labor productivity and retention. By asking workers for their opinions, firms can learn information about worker preferences, production process improvements, and other insights. However, asking workers about their opinions may create the risk of disappointment by changing workers’ reference points. If firms cannot follow through on workers’ suggestions, workers may feel a loss and lower their productivity in response. It is difficult to identify a disappointment effect of worker voice because firm responses to workers’ suggestions are difficult to randomize. I employ an online experiment that randomizes the solicitation of worker input and employer response and uses real effort tasks to test the effect of worker voice with and without follow-through.

External Link(s)

Registration Citation

Citation

Ho, Helen. 2022. "Worker Voice and Disappointment." AEA RCT Registry. January 20. https://doi.org/10.1257/rct.8816-1.2000000000000002

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

The experiment randomizes whether respondents are asked for their preference over two types of tasks and randomizes which task respondents receive.

Intervention Start Date

2022-01-17

Intervention End Date

2022-01-21

Primary Outcomes

Primary Outcomes (end points)

Work quality and willingness-to-work

Primary Outcomes (explanation)

Work quality will be measured as the Jaro-Winkler distance between the respondent's transcription and the correct characters. Willingness-to-work will be elicited using a Becker-Degroot-Marschak mechanism.

Secondary Outcomes

Secondary Outcomes (end points)

The respondent's views of the researcher's leniency, voice-seeking, and responsiveness to voice.

Secondary Outcomes (explanation)

Experimental Design

I will recruit participants on an online platform to complete transcription tasks.

Experimental Design Details

I will recruit participants on an online platform to complete transcription tasks. In the first session, respondents will receive both English and CAPTCHA tasks so they can form preferences. At the end of the first session, respondents will be told that they will be assigned one type for the second session. Respondents are then randomized between the voice and no voice treatments. In the second session, respondents are randomized into the English and CAPTCHA task types. Each respondent must complete a set number of the assigned task type. They are also given a chance to complete additional tasks of the same type for bonus pay through a Becker-Degroot-Marschak mechanism.

Randomization Method

Random number generator within Qualtrics

Randomization Unit

Individual-level randomization. Outcomes are at the individual-task level (quality) and at the individual-bonus level (willingness-to-work).

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

1220

Sample size: planned number of observations

9,760 individual-tasks for the quality outcome. 6,100 individual-bonus levels for the willingness-to-work outcome.

Sample size (or number of clusters) by treatment arms

Voice-CAPTCHA: 540
Voice-English: 60
No Voice-CAPTCHA: 540
No Voice-English: 60

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Work quality in Jaro-Winkler distance (continuous 0 to 1): mean = .96, SD = .06, ICC = .04, MDES = .004 in the distance measure, or 0.4 percent of mean Willingness-to-work (number of tasks between 0 and 100): mean = 24.8, SD= 19.3, ICC = .7, MDES = 3.1 tasks, or 12.5 percent of mean

Supporting Documents and Materials

IRB