Bayesian Network for Survey Satisficing Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Surveys are often skewed by 'satisficing' behavior, where respondents provide half-hearted or arbitrary answers, leading to inaccurate data that diminishes the effectiveness of survey results.

Innovation Solution

The system automatically detects satisficing in survey responses by learning a custom probabilistic model from questionnaire data, calculating a satisficing score for each response, and using a Bayesian network to identify outliers based on question interdependencies and conditional probabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated surveys are conducted without human intervention, then survey efficiency and speed are improved, but data accuracy deteriorates due to satisficing behavior

Engineering Contradiction:

Improvesurvey efficiencyVSAvoiddata accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system implements feedback by analyzing survey response patterns and using probabilistic models to identify satisficing behavior. The Bayesian network continuously evaluates responses against learned patterns from training data, providing feedback on response quality and enabling automated detection of satisficing participants without human intervention.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent replaces manual human review of survey responses with an automated computational system. The mechanical process of human data validation is substituted with probabilistic modeling and Bayesian inference algorithms that automatically detect satisficing behavior through pattern recognition in response data.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Device complexity

If traditional survey analysis methods are used, then implementation simplicity is maintained, but detection capability of satisficing responses is insufficient

Engineering Contradiction:

Improveanalysis method simplicityVSAvoidsatisficing detection capability

Core Design Contradiction:

Device complexityVSDifficulty of detecting and measuring

Solution Approach 1:

The system performs preliminary action by training probabilistic models and Bayesian networks on survey data before actual survey analysis. The system learns patterns of satisficing behavior in advance from training datasets, enabling it to automatically detect and flag problematic responses during production survey analysis without requiring complex manual intervention.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS8583568B2Systems and methods for detection of satisficing in surveys

Publication Date: 2013.11.12 SURVEYMONKEY

AI summary

Response data relating to a plurality of responses to a survey is received. The survey comprises a plurality of questions. The response data for each of the plurality of responses comprises a plurality of answers to at least a subset of the plurality of questions. A questionnaire response model is created using the response data. It is then determined, for each questionnaire response, a respective probability that the respective questionnaire response represents satisficing, such that where the respective probability exceeds a threshold, the respective questionnaire response is identified as an outlier, and where the respective probability does not exceed the threshold, the respective questionnaire response is identified as an inlier. A representation of each questionnaire response is output, each respective representation reflecting the likelihood that the respective response to which the respective representation represents satisficing.