Attorney-Case Matching: Evidence from Online Legal Marketplaces

Last registered on January 23, 2023

View Trial History

Pre-Trial

Trial Information

General Information

Title

Attorney-Case Matching: Evidence from Online Legal Marketplaces

RCT ID

AEARCTR-0009699

Initial registration date

January 14, 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

January 23, 2023, 6:36 AM EST

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

Locations

Country

United States of America

Region

Primary Investigator

Name

George Vojta

Affiliation

University of Chicago

Contact Primary Investigator

Other Primary Investigator(s)

Additional Trial Information

Status

In development

Start date

2023-01-09

End date

2023-10-01

Keywords

Crime, Violence, & Conflict, Gender, Labor, Welfare

Additional Keywords

JEL code(s)

K00, K13, K15, K40, K41

Secondary IDs

Prior work

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

Abstract

I will conduct a case-posting correspondence study on an online legal marketplace where people posts their cases and lawyers bid to represent them. This study will randomly assign race, gender, income, and prior arrests to a large number of fictitious case postings. I will study differences in the lawyer responses to determine whether and to what extent lawyers discriminate when it comes to deciding who to represent.

External Link(s)

Registration Citation

Citation

Vojta, George. 2023. "Attorney-Case Matching: Evidence from Online Legal Marketplaces." AEA RCT Registry. January 23. https://doi.org/10.1257/rct.9699-1.0

Sponsors & Partners

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

Request Information

Experimental Details

Interventions

Intervention(s)

Intervention Start Date

2023-01-09

Intervention End Date

2023-10-01

Primary Outcomes

Primary Outcomes (end points)

There are 5 different outcomes I will be analyzing as part of the study. Lawyers on the platform submit bids and thus I want to see how the components of the bids and how the lawyers themselves change with these variables. The outcomes I want to analyze are:
1. Contact Rate
2. Number of bids
3. Average fee (and potentially fee structure)
4. Lawyer quality (both by stars and law school ranking)
5. Speed of Responses

Primary Outcomes (explanation)

Average fee will be standardized whether it is hourly or flat fee. Lawyer quality will use both the star system LegalMatch has in place as well as taking the law school ranking of their education.

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

The study is a correspondence-style audit study in which I will submit fake cases to online legal market places. The intervention will be varying key components of the hypothetical cases. Cases will be submitted over the course of a year to different zip codes throughout the country.

Experimental Design Details

The study is a correspondence-style audit study in which I will submit fake cases to online legal market places (LegalMatch and UpCounsel). The intervention will be varying key components of the hypothetical cases. Cases will be submitted over the course of a year to different zip codes throughout the country.

There are two wings to the study: first is personal injury, second is criminal law. Every week we submit 1 case of each type to 120 metro areas around the USA. Cases are submitted from 120 generated accounts that are stratified on race, gender, and racial sound names (15 of each types, 8 groups). Each metro area will receive 1 case of each type from one of these 120 accounts in a random order. Other characteristics are randomized. These characteristics include: age, income, zip code option (picks one of the top 5 most populous), bail amount, prior convictions, and how many days ago the incident happenen.

Randomization Method

Randomization done by python random number generator.

Randomization Unit

Individual level after stratification

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

I will submit approximately 6,000 cases over the course of half of a year.

Sample size: planned number of observations

6000 cases, number of lawyer responses is unknown.

Sample size (or number of clusters) by treatment arms

750 per race-gender-racial_name block. Other characteristics will be added in structurally.

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

By looking at existing means and variances on test data, I should be fine with about 300 per cell, without the linear terms from age, bail, etc. At 8 cells this would be about 2,400 observations.

Supporting Documents and Materials

IRB