Peer Review

Bias & Fairness in Employee Promotion

Material Detail page

Reviewed:: Apr 20, 2026 by Business

Bookmark

Ratings

Overall Rating:

Content Quality:

Effectiveness:

Ease of Use:

Overview:

This interactive tool lets you design a simple promotion algorithm and watch how it affects different demographic groups. The authour worked with a synthetic company of 240 employees who needed evaluation for promotion. The dataset reflects a common problem, indicating that, even when people have similar talent and performance, they often receive different levels of support, training, and recognition based on their demographic characteristics. These historical patterns get baked into the data that algorithms use to make decisions. In this study, how much weight the algorithm should place on different factors like performance scores, manager ratings, and training hours were discussed. Adjusting these weights will result in how promotion rates shift across gender and racial groups. Furthermore, some weight combinations will amplify historical biases, while others might reduce disparities. The goal is to understand that algorithmic fairness isn't automatic. When we feed biased historical data into neutral-looking formulas, we often obtain biased outcomes. The challenge results in balancing predictive accuracy, business needs, and equity concerns.

Type of Material:

Simulation

Recommended Uses:

Could be used in group work or discussion
Could be used as homework
Could be used for presentation.
Could be used in online class as group assignment where students pretend to be part of a promotional board.

Technical Requirements:

Internet Browser such as Fire Fox, Chrome or Microsoft Edge

Identify Major Learning Goals:

Students will explain how biased historical data can impact algorithms that optimize predictive accuracy
Students need to understand that fairness is not just about algorithmic neutrality.
Students will learn that the disparate impact ratio will help in quantifying the differences.
Students will understand how organizations audit historical data for bias. Also, the ethical implications of using AI systems that learn from biased human decisions will be learnt.
Students will learn that HR systems involve far more complex, legal considerations, and stakeholder interests, and will be able to use this tool as a guide.

Target Student Population:

Professional and graduate school students enrolled in human resources and organizational behavior programs.

Prerequisite Knowledge or Skills:

A introductory course in business analytics; knowledge of how algorithms are used in human resources, institutional bias, disparate impact ratio

Content Quality

Rating:

Strengths:

Good and adequate examples are used.
Academic definitions are used along with appropriate learning exercises
Good command of english language
Tool is timely especially with the increased use of AI in promotions and hiring and the assumptions that algorithms lack human bias
The instructions for use were clear.
The different factors had explanatory notes to accompany them.

Concerns:

The learning material provides an overview of how different promotion factors impact gender and racial groups. However, demonstrating how the different promtion factor impacts subgroups such as a female from race A, a female from race B, and a female from race C would reveal more detailed information regarding disparities.
There was no references cited or background information provided to explain how the tool was developed and its predictive accuracy.

Potential Effectiveness as a Teaching Tool

Rating:

Strengths:

The algorithm settings are well-explained
The learning objectives well explained and practically engaged
The summary followed a logical sequence with good meaning
The disparities are well pointed out
The tool demonstrates relationships betweeen promotional factors and the groups that are impacted by them.
It can easily be integrated into curriculum assignments.

Concerns:

Pre-requisite knowledge is needed to understand the concepts introduced by tool and its accompanying questions.
The tools or accompanying information does not build upon prior concepts or reinfornce concepts progressively.
It also can not be used to meausre student learning outcomes.

Ease of Use for Both Students and Faculty

Rating:

Strengths:

Great access to the material used
Consistent and visible control of the topic
Well explained class usage, with clear instructions
Instructions are provided for classroom use.
The tool is intuitive.
It has sliding bars to change its values.
The tool is engaging and interactive.

Concerns:

None

Other Issues and Comments:: Well-explained skills in summary and comprehension understanding
Creative Commons: