The 5DGM Journey

Welcome to the story of the 5DGM assessment. It tells of how it came about and why it is what it is today. We hope you enjoy the narrative...

Written by Ian Dyason, Founder & Developer of the 5DGM

It all started on 26 September 2012.

My previous company, A I Training & Consulting, hosted a visit by Prof Jeanne Liedtka from Darden Business School, UVA. We organised a workshop entitled "Growing From Within" (pictured above) This is the precursor of the design thinking programs that Jeanne would later go on to champion.

In this workshop, Jeanne shared her research into companies that were successful in growing their business organically and she introduced to us the concept of the growth and fixed mindset. This was the first time any of us had heard of this concept, even though it had been published 6 years prior by Prof Carol Dweck. According to Jeanne, companies that were successful had Growth Catalysts to push innovation and market expansion. The common ground for all these individuals was that they were powered by the growth mindset.

When I asked Jeanne how she identified these individuals' mindset and how she measured it, her response was that there was no proper assessment tool to measure one's mindset at that time, and so she used the DISC profile as proxy. She said that the growth mindset coincided with the D-I profile (a profile that I incidentally also possess!)

And that became how we first deployed the DISC assessment for participants of our Growing From Within program, which we conducted for several organisations in Singapore and Southeast Asia.

However, it quickly became clear that the DISC tool was too blunt an instrument, and it did not focus on the right behaviours for the growth mindset. Realising that there was no objective, behaviour-based growth mindset assessment in the market, I decided to build it ourselves.

That was the start of a journey that would take more than 10 years to fulfill and countless hours of research.

The Initial Premise

Our initial premise was guided by Dweck's work; that is, growth was good and fixed was bad. It was a binary concept that would define whether one would be able to add value to an organisation or not.

Our first hypothesis was:

"That we could create behaviour statements aligned with the fixed and growth concepts, and we could identify a person's mindset using these statements."

This was harder than it seemed. Dweck's 2-set, 4-statement assessments (one set on learning and one set on personal qualities) was too brief. Her assessment asked people to agree or disagree with each statement, and depending on their response, it would determine if the person to had a growth or fixed mindset.

We found 4 statements to be inconclusive because what would it mean if a person had a growth response for two statements and a fixed response to the other two?

We ultimately came up with 9 statements, each of which was similarly constructed as a yes/no response. The statements were:

Life is a journey of learning
I embrace uncertainty
I seek new experiences
I have broad skills
Customers are data to learn from
I manage risk through analysis
I place big bets slowly
I fail more often in new situations
I avoid looking foolish

Some of these statements were related to the fixed mindset and some to the growth. So long as everyone answered these truthfully, we could plot whether the person had a fixed mindset or not.

We were now ready to start collecting data.

400 People Cannot Be Wrong!

We initially invited more than 30 people who were known to us to take this assessment without telling them what this was about. After we had collected their responses, we sat down with them to explain the findings and to ask deeper questions. However, 30 people was not enough for us to create a baseline so we asked friends of friends to join in. We ended up with more than 400 respondents! What we discovered blew up in the face of the dominant logic of the growth mindset:

people can have BOTH fixed and growth mindsets!

Surely 400 people cannot be wrong (assuming, of course, they did not game the system by saying yes when it was no, and vice versa)?

In searching for an answer to this dilemma, we postulated that "mindset" has many facets rather than a single dimension. As such, our next hypothesis was:

"There is more than one dimension to the growth mindset and we can determine them objectively."

We applied systems thinking and force field analysis on the concept of "mindset", brainstorming all the different behaviours that drive mindset, be it fixed or growth. What resulted was a spaghetti diagram linking the many drivers together. But what we ultimately discovered was that most of the interconnections came through five main drivers, and these became the "centres of gravity" of the mindset system. These centres of gravity are learning, resilience (bounce back ability), risk taking, discovery (the propensity to move forward despite not knowing all the outcomes) and conclusion (the pride of past achievements causing one to jump to conclusions). While interconnected, these centres of gravity, also known as dimensions, uniquely drive a person to act in a certain manner. And, more importantly, one could be fixed in one dimension and growth in another.

This, therefore, explained how a person could be both fixed AND growth at the same time, debunking the myth of the dominant logic!

We now had the beginnings of an alternative theory of the growth mindset and we had the behaviours that contributed to it.

Measuring Mindset

Another major milestone in our discovery of the mindset was its measurement. Our third hypothesis was:

"We can measure mindset and it is valid."

With the newly-discovered 5 Dimensions and the behaviour statements linked to each in hand, we realised that our rudimentary 9-statement assessment wouldn't do. Indeed, we were now in the position to make our questionnaire more sophisticated. We created 40 behaviour statements (8 statements per dimension) in an Excel spreadsheet and asked individuals to rate themselves on a scale of 1 to 7, where "1" was "Not at all like me" and "7" was "Totally like me". Our premise was, given the law of large numbers, 50% of the respondents would have an average score below 4.0 (the mid-point between 1 and 7) and 50% of respondents will be above 4.0.

That was not to be!

Not only were more people below 4.0, but if we wanted to fit the 50-50 split between fixed and growth, the score would be about 2.4. Furthermore, we knew many of the respondents personally, and we knew them to be growth minded, yet their scores were below 4.0!

Obviously, we now know that this is to be expected because people may relate to different behaviour statements differently, and so objectively splitting the score down the middle to distinguish between growth and fixed doesn't make sense. But those were early days in our research and this fact was not very apparent to us then. We had wanted to ensure that our hypotheses aligned with the dominant logic. Unfortunately, we had a hard time doing that. The data simply could not fit the dominant logic that growth and fixed was a binary concept!

The only way one could make sense of the data was if there was a middle zone, where one could have both fixed and growth tendencies, and they displayed these tendencies based on the prevailing context. Indeed, when we split the measurements in this manner, all scores made sense!

What we had discovered was that the growth and fixed mindset was indeed not a binary concept; and one's mindset lied along a continuum where the two ends were the fixed and growth zones, and the middle was the zone where one could display both fixed and growth tendencies. We called this the Transition zone, since one could transit between fixed and growth tendencies, depending on context!

This, yet again, flew in the face of the dominant logic!

Growth is not necessarily good...

Another dominant logic sacred cow that we "brought to pasture" was the notion that growth was good and fixed was bad. This concept challenged me when someone who was obviously very successful in his career was found to be fixed minded - and he was very proud of it! It turned out that the person was in procurement and had to ensure product quality, price, quantity and delivery schedules were all aligned. He was fixed in the way he conducted his affairs, but that was not a bad thing at all, since that was required of the job role!

The dominant logic had painted that success came on the back of the growth mindset, but here was an example to the contrary! It proved that growth was not necessarily good and fixed was not necessarily bad. Again, it all boiled down to context!

Systematising

By now it was 2015.

We had taken more than 3 years to come to the above conclusions and we had been using Excel to get respondents to answer the questionnaire and to obtain a radial graph of their profile. But that was not helpful for meta-analysis (we didn't have Copilot and PowerBI then). It was also not professional when we wanted to deploy that in training. There was now a need to systematise.

In this early stage, we partnered with an external assessment provider who had their own platform for collecting inputs and then creating the growth mindset output. To distinguish between the outputs for his own assessment and our outputs, we named ours the 5 Dimensions of the Growth Mindset (5DGM). This was meant to be a "placeholder" name but the name has stuck till now. (But that does not mean it is a sacred cow, and that we won't change it! However, that's a discussion for another day!) We went into agreement where GCA would own the foreground IP for the 5DGM and the assessment platform owner owned the background IP. (We later saw a disadvantage to this because we did not own the data, and hence could not perform the meta-analysis as we had earlier wanted to do. There would also be data security issues that would later have to be solved.)

We created a report for the output so that after the respondent took the assessment on the platform, they would get something to read. We ran this partnership for 4 years and used the output for several of our programs. However, in 2019, following a disagreement with the new partner of the platform owner, we decided to part ways.

Suddenly, we were out of a platform but had customer requirements to fulfill!

It was time for us to create our OWN platform!

Version 1

We created Version 1 with Version 2 in mind. But we did that because we needed to get the technicals correct and to ensure that we continued to have a market. Hence, we kept our proof of concept (POC) as cheap as possible.

We created a simple site using the ipsative assessment model. An ipsative model is one where you rate yourself on a linear scale, like the one we developed on the Excel spreadsheet. In essence, we took the algorithm from the Excel model and put that on an online platform. Our intention was to meet customer requirements while we sharpened our technical capabilities.

We knew people were gaming the output based on their preconceived notions of what was right. While we exhorted them to be truthful, we knew that we could do nothing if they opted to be more "flexible" with their responses.

As such, the minute we had tested all our capabilities, and understood where the market was, we created Version 2.

Version 2

This was a much more sophisticated model with an output that could not be manipulated. Using the forced-choice assessment model, each respondent was given 40-sets of 3 behavioural statements each, and they had to rank order the statements from "most like them" to "least like them" by physically moving the statements up or down onscreen so that they could see the final order before locking in their response. Because the algorithm was secret and complex, respondents wouldn't know the logic and therefore could not work towards an "ideal" answer. As such, the results of this assessment were highly valid (meaning it actually measures what it says it measures) and, due to the construct of the output, the system was also very reliable (meaning, it will produce the same output given the same inputs over time). This is what we have today.

Yet, to have gotten here required more research.

Scholars have published that the forced-choice method is the best method for assessments, and one could use a two-statement, three-statement or four-statement set. We decided not to try the two-statement sets because we knew this could be somewhat gamed (although research said otherwise). So we focussed on the three-statement and four-statement sets, and tried them out with volunteers. While I would have loved to use the four-statement set, volunteers reported mental fatigue in trying to rank order four statements. And if fatigue set in quickly, they would probably not have been able to complete 30 sets of such statements, making the output less valid. The three-statement sets were less mentally fatiguing, but due to the 5 dimensions that we had to assess, we could not do it with any less than 40 sets. So this was what we settled on.

Other Things We Could Do

Once we had a valid and reliable platform, we could do many things with it.

We could now predict future behaviour since we were confident with the inputs received. By creating "Optimal Profiles" for different job roles, we could overlay each individual's mindset profile onto it and determine the "gap" between their profile and the optimal. These gap analyses would also provide developmental suggestions, thereby making the reports more complete. This allowed us to create assessment reports for leaders, entrepreneurs, solopreneurs. We also have reports for career success and sustainability. So long as we could identify the behaviours to drive a successful mindset for these roles, we could create a targetted report with specific development suggestions hitherto unavailable!

This also allowed us to customise the reports for different target audiences, especially in different countries, for different languages.

We currently have the assessment platform in English and Vietnamese, with Spanish and Chinese language being planned for 2026.

This also provided us the opportunity to install the assessment in-house with large organisations. We currently have specialist reports for organisations in the fast food & FMCG space.

2026 promises to be a very exciting year indeed and we cannot be more proud of the journey we have traversed over the past 14 years!