Digital Screening Platform for Bias-Free Data Quality

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing data quality checks in online research platforms are ineffective in ensuring high-quality data while maintaining participant diversity, often leading to biased samples due to their reliance on cognitive taxing measures that favor educated and socioeconomically privileged individuals, and fail to accurately assess attentiveness and non-random responding.

Innovation Solution

A digital screening platform that uses a tailored assessment with a self-replenishing bank of questions created by a trained algorithm to evaluate attention levels, language proficiency, effortful responding, and response validity, incorporating security checks, language proficiency engines, and association question engines to generate questions that adapt to individual participant performance, ensuring a balanced and accurate selection of high-quality data without bias.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional data quality checks are used to screen participants, then data quality is improved, but participant diversity deteriorates due to bias against underrepresented groups

Engineering Contradiction:

Improvedata qualityVSAvoidparticipant diversity

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent changes the parameters of screening questions from cognitively demanding formats (multiple choice, complex instructions) to simpler formats (open-ended association questions) that reduce bias while maintaining data quality detection capability. This parameter change allows the screening to be more inclusive of diverse participants while still effectively identifying inattentive respondents.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent uses a large bank of disposable screening questions that can be randomly selected and regenerated. Instead of relying on a fixed set of questions that may be biased, the system uses many different question instances that can be discarded after use, reducing the impact of any single biased question on overall screening fairness and diversity.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

2Measurement precision

If cognitive taxing measures are used in screening, then attentiveness detection is improved, but bias against less educated individuals increases

Engineering Contradiction:

Improveattentiveness detectionVSAvoideducational equity

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

Instead of using complex questions that require high cognitive ability to detect inattentiveness, the patent inverts the approach by using simple association questions where the expected response time is short. This inversion allows detection of inattentiveness through response pattern analysis rather than through cognitively demanding tasks, thereby reducing educational bias while maintaining measurement precision.

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent replaces mechanical cognitive tasks (reading comprehension, logical reasoning) with a different detection mechanism based on response time analysis and association strength measurement. This substitution uses computational analysis of simple responses rather than requiring participants to perform cognitively taxing operations, thereby reducing bias against less educated individuals.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If simple screening questions are used, then participant accessibility is improved, but accuracy in identifying inattentive participants deteriorates

Engineering Contradiction:

Improveparticipant accessibilityVSAvoidinattentiveness detection accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent uses a large bank of screening questions that can be continuously drawn from, ensuring that simple accessible questions are always available. The system maintains accuracy by using multiple questions from the bank rather than relying on a single simple question, thereby preserving inattentiveness detection accuracy while maintaining participant accessibility through simple question formats.

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

The patent creates multiple copies of simple association question formats with different word pairs and associations. Instead of using one complex question, the system uses many simple question copies that collectively provide accurate inattentiveness detection through aggregated response pattern analysis, maintaining both accessibility and precision.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS11227298B2Digital screening platform with open-ended association questions and precision threshold adjustment

Publication Date: 2022.01.18 PRIME RES SOLUTIONS LLC

AI summary

Systems and methods for tuning a digital screen to provide high quality data are provided. Methods include determining a target level of participant data quality associated with accurate completion of an online survey, determining a participant screening threshold based on the target level of participant data quality, and adjusting a survey screen based on the participant screening threshold. Methods may achieve high data quality without sacrificing participant diversity. Methods may also include transmitting the survey screen to a computing device associated with a participant, and receiving a response of the participant to the survey screen on the computing device. When the response fails to achieve a predetermined threshold response, methods may include rejecting the participant from the survey.