Blog | The LenddoEFL Assessment Part 2: Measuring how people answer questions with metadata

By: Jonathan Winkle, Manager of Behavioral Sciences, LenddoEFL

The last post showed how our psychometric content reveals people’s personality traits, but our assessment also captures an abundance of metadata. Metadata is information about how people process the questions and exercises they complete. Here are some examples.

• How long did an applicant take to answer a question compared to their average response time?

• How many times did an applicant change their mind and switch their response before submitting their answer?

• Is the applicant’s information consistent with their written request to the financial institution? (e.g., requested loan amount)

By measuring metadata, LenddoEFL’s approach goes beyond what is possible in traditional credit applications to reveal more information about applicants. Consider the following question from our test:

Which BLUE person is most like you?

For this question, we consider how long it took the applicant to slide to one answer or another and whether they changed their opinions in the middle. Someone who is confident that they are an organized person should move the slider in only one direction and relatively quickly. Quick, smooth answers belie confidence, whereas slow, wavering responses demonstrate uncertainty.

The relationship between response time and default rate can be complex. Consider another psychometric exercise:

In this case response time was a non-linear predictor of default, where both slow and fast response times were associated with a greater credit risk!

There are many ways to interpret response time metadata. If an applicant answers a question quickly, are they confident or are they cheating? If they are taking a long time to respond, are they having difficulty understanding the question or putting extra effort into getting their answer right? By collecting metadata across all questions, we can compare a single response time to the applicant’s overall response time distribution to differentiate things like confidence and cheating (see graph below).

An example distribution of response times generated from artificial data

Conclusion

Metadata reveals another layer of behavior on top of the personality traits we target and can be used to identify features such as confidence, cheating, and confusion. These behavioral traits can be used for predicting default and ensuring that we are collecting high quality data for our models.

Lodex Blog | The Future of Data-Driven Financial Inclusion Posted by Aisha Hillary-Morgan

"In Australia, millions of people find themselves in a chicken-or-egg-type dilemma when it comes to getting credit. Even though they have steady income, they still can’t access credit because of lack a formal credit history. Yet, most of these consumers carry a smartphone, are online and connected through social networks, leaving behind a digital footprint that can be analyzed to better understand who they are and their attitudes toward credit.

This is why we have teamed up with LenddoEFL, the leading technology platform powering data driven decisions for financial services, to help them create more of a credit story. Your Social Score will use, with your consent, your digital footprint to provide additional insights for borrowers and for lenders to more efficiently make a preliminary assessment." Read the full article

Originally posted by our partner Lodex

Reaction from Mynt: "this merger will set new standards in the industry"

Thanks to our partners at Mynt for sharing their feedback on the merger.

"In a country where native bureaus only have data on less than 5% of the population, the alternative credit scoring methods and technologies Lenddo and EFL have designed and created have enabled us to work towards financial inclusion in the Philippines, from consumers to MSMEs. With the work we've done with Lenddo for our consumers products and with EFL to assess businesses, we can testify that these two companies have many synergies and complementary competencies. With the combination of their knowledge and models, and the combination of social, telco, and psychometric data, we have no doubt that the company resulting from this merger will set new standards in the industry."
-- Anthony Thomas, CEO, Mynt (Globe Fintech Innovations)
 

"With 70% of Filipinos remaining unbanked and with less than 5% in credit card penetration, credit data about the majority is virtually non-existent. This creates a cycle where most Filipinos remain financially underserved. However, with the emergence of technology for alternative credit scoring methods, we are enabled to work towards financial inclusion in the Philippines. With the work we've done with Lenddo to credit score consumers and with EFL to evaluate businesses, it comes as no surprise that a merger between these two innovators is happening. With the experiences and learnings both have acquired separately, we are excited to see the birth of even better and richer data with the marriage of these companies."
-- Jean-Francois Darré, Chief Analytics and Risk Officer, Mynt (Globe Fintech Innovations)