Government Operational Research Service

What you need to do

You need to answer four questions. You have up to 300 words for each of
your responses. Each question refers to a competency.

Use the next page to answer the questions, when you’re finished you can
copy and paste the content into the response box.

Problem Structuring

Give evidence, from work or study, of where you applied a problem

structuring approach.

Problem structuring is an important and early step in analysis. It is the

process of breaking down a complex problem, into a clear structure and
formulating specific questions that could be tackled through analysis of
available data and or numerical modelling.

Please describe:
• • Your role;
• • What you did to identify and clarify any issues relevant to the
problem;
• • How you organised and presented your thinking during the
process.

Quantitative Analysis

Give evidence of where you have used quantitative methods to tackle and
solve a problem.

This competency refers to a situation where the analytical questions have

been established and you now need to produce a concrete and quantified
solution using numerical analysis and/or modelling. You may choose to
cover the analysis part of the same project you covered in the previous
question, but this is not essential.

Please describe:
• • Your role;
• • Any analytical techniques you used and how you used them;
• • How you used available data or produced it;
• • How you assessed the accuracy and usefulness of your
results.

Delivering Results
Describe a challenging project or activity which you have planned and
taken through to a conclusion.

It is important that OR analysts plan their work clearly and can respond to
clients requests within deadlines, being proactive and using initiative
when problems arise.

Please describe:
• • Your role in the project;
• • Your objective and the steps you took to plan and monitor the
project;
• • How you dealt with challenges, both foreseen and
unexpected.

Engaging People

Describe a situation where the relationships you have built, either within a
team or with customers, have helped you to achieve a shared goal.

The ability to build productive relationships is important as it allows you to

achieve goals in many different situations, whether engaging with peers
or senior colleagues.

Please describe:
• • Your role within the context and goal;
• • How you worked with any team members, customers and
interested parties;
• • How you dealt with challenges;
• • The impact you had on others.

Do you currently work for GORS? If so, in which department?

Copy and paste the headings and your responses into the
response box.

Problem Structuring

In my first and second years of university, I worked for a nonprofit social entrepreneurship
startup called Share Meals (sharemeals.org), whose mission it was to leverage technology
(amongst other solutions) to end food insecurity, i.e., hunger, on campus.

When I joined, we were too small a group to give formal titles to our members, but I essentially
worked on the prong of the initiative that was dealing with a way to demonstrate the
underutilisation of university meal plans and. This was a complex and multifaceted problem that
required a thorough review and analysis of all the variables before being acted upon.

I knew that I was not the only one who had surplus meal swipes on their plan as I had spoken to
plenty of hallmates in my dorm who were in a similar boat. However, in order to make it clear to
the university administration that this problem was widespread, I proposed running a survey to
gauge the actual level of underutilization (and the concomitant “sunk cost,” tabulated by
multiplying the number of unused meals by the cost per meal). This turned out quite
successfully, and even though we only got five hundred responses to the survey, the sunk cost
figure was enormous (above a quarter million dollars).

The organisation and presentation of this idea was a nontrivial undertaking. Our small group
worked tirelessly on the survey, and I suggested adding humorous elements (e.g., a feedback
report that would show each responder how long their remaining meal swipes could feed The
Rock (i.e., Dwayne Johnson) for) in order to increase the probability of potential responders
taking the time to complete the survey. This worked well, and overall the work I did on problem
structuring was quite successful to the endeavour of the group as a whole.

Quantitative Analysis

In the context of working as a data scientist at the financial technology firm Biz2Credit in New
York in the fall of 2018, I had several largely quantitative projects. One particular one was in a
client project wherein the client wanted an “industry heatmap” showing the number of firms by
industry in their own geographical footprint (as to better tailor their business to local clients).

My role was to fulfil this request, using any resources I could find. As the client’s footprint was
largely in New York and South Florida, I set off to find data from the United States Census
Bureau’s Statistics of U.S. Businesses division. Large data files provided the raw data for firms
across the country by industry, but a nontrivial amount of additional work was required to parse
this data coherently and turn it into something useful for the client.

I used a combination of techniques to streamline the data, from the Excel VLOOKUP function to
match zip codes to cities, some basic mathematical formulae to tabulate percentages of
industries by region, and some visual modeling to make a color-coded “heatmap” of the
concentration of various industries by region. Fortunately, I did not have to “produce” any data
as it was all provided by the Census Bureau, but the provided raw data was quite unintuitive and
needed significant “cleaning up” and presentation before it could be understood easily.

I concluded my analysis of the data with some neat tables of industry percentages alongside
heatmaps, and I assessed the accuracy and usefulness of my data by considering how the client
would specifically be able to use the numbers provided to tailor their products and services to
local clients better. This yielded some useful ideas.

Delivering Results

When I was working as a tutor and test statistics coordinator for an intensive summer
residential SAT test prep camp in the summer of 2016, my cohort of fellow teachers and I
noticed a problem with the breakdown of mathematics classes. While we had originally grouped
our forty students into four groups of ten based on their raw scores in the diagnostic SAT exam
they took prior to starting the camp, it became clear after a week of classes that ability levels
within the classes (far more so in mathematics than in reading or writing) differed greatly. As
such, it was my duty (as the person in charge of grading the tests, and consequently, ensuring
how the ability groups might be (re)structures) to figure out a solution to this discrepancy.

I proposed overhauling the rather simple algorithm we had for assigning different students to
different ability groups with a new one based on concept proficiency rather than raw score. This
way, all the students who, for example, needed help with linear equations or inequalities, could
be grouped together such that the teacher of that class could proceed without leaving any
student behind or making other students feel bored.

The idea was well-received. The biggest challenge was the upfront labor. We could not simply
use machine learning to classify the mathematics problems in our weekly SAT practice tests into
all the various designated categories—this would have to be done by hand. However, I argued
that the payoff would be long-term, so the entire staff spent a day meticulously parsing the
practice tests by mathematical concept, and at the end, I integrated these findings into the
software used for test scoring (Microsoft Power BI), using it to effectively restructure the ability
groups, thus obviating the problem of concept proficiency discrepancies.

Engaging People

When I was the Senior Patrol Leader of my Boy Scout Troop in California in 2015, I was
responsible for running a troop of 100+ scouts and 30+ adult leaders. My role included
overseeing the meetings of the Troop Leadership Council to plan camping trips and outings,
presiding over the weekly troop meetings, liaising with adult leaders and serving on the long-
term planning committee, and organising all the logistics for our annual troop-wide weeklong
camping trip.

Needless to say, the position required manifold skills. One skill I learnt to use very was
delegation: it was certainly impossible for me to singlehandedly juggle all the various tasks of
running the troop, so it was up to me to build close and trusting relationships with members of
the Troop Leadership Council such that I could delegate tasks to them effectively. I did this not
only by speaking to them professionally and allowing them an appropriate share of the
responsibility but also by spending time with them during camping trips and outings, playing
card games with them during recreational time, pitching tents with them, etc. This sort of
camaraderie made it easier to lead the troop smoothly.

Challenges certainly arose with troublemaker scouts and even protective parents reluctant to
send their scouts on camping trips without accompanying them themselves. I handled
situations like the former with a mix of diplomacy, delegation, and friendliness, and situations
like the latter with being open and available to parents as much as I could to give them a sense
of the troop’s workings.

At the end of my six-month tenure, coinciding with my graduation from high school, my Troop
Leadership Council was extremely well-prepared to continue the successful operations of the
troop and my second-in-command won the following senior patrol leader election by a
landslide.