Diary of a data project - Part 2
Toni SekinahDataIQ has partnered up with Tech Nation UK to find out what takes place when interrogating a dataset to find insights. With entry number two, Diana Akanho takes us through the second stage of a data analysis and presentation project, with a recap of the planning stage and an explanation of the doing stage.
Here’s the second entry of Diana Akanho, senior insights manager at Tech Nation.
This project can be broken down into four stages: planning, doing, reviewing and execution.
The ‘doing’ stage is when the main bulk of the analysis takes place. Reviewing and execution are the final stages of the data project and this involves, checking analysis, getting feedback from stakeholders - both internal and external. So, your execution will be in regards to how you want to deliver your insights, whether that be through a report and presentation, or a launch event.
The planning stage took two weeks. The main obstacle in that phase, which was overcome, was deciding what tool to use in order to handle ‘big data’.
In terms of the hypothesis, we had the belief that demand for roles in the digital tech sector was increasing; this is proving to be true.
I saw that patterns became apparent with the type of roles that offered the highest median salaries, and the locations where there was a bigger demand for digital tech roles.
Beyond this, we wanted to explore different themes, such as employers demand for skills, roles and the changing nature of these across different locations within the UK.
My work process when doing data analysis is to initially get the summary stats to understand the overall picture of the data I am going to analyse. When I have gathered the insights, I think about how I want to visualise the data.
I am starting to get an idea of the conclusions I will reach. As I am analysing the data and understanding the insights, a picture is starting to come into focus. As a result of the insights, the title of the report was formed, ‘A Bright Tech Future’.
Whilst analysing the data, a few further questions arose.
- How do roles advertised in particular sectors compare to others?
- How has this changed over time?
- How has employer demand shifted for specific types of roles and in turn is this reflected in the salary, hence, showing the value of these roles?
We did end up looking at the other data sources such as the Labour Force Survey and the Annual Business Survey. George Windsor, head of Insights at Tech Nation, was responsible for this.
George said: ’Labour markets are incredibly complex, and we wanted to unravel what they look like both nationally, and regionally. To do this, we had to use a range of different data, including survey based data from the Office for National Statistics. We used the Annual Population Survey, waves 2-4 in 2017, and wave 1 in 2018 - to understand people working in digital tech jobs across the UK economy. We also used the Interdepartmental Business Register, which includes data on all companies in the UK that are PAYE or VAT registered, to look at people working in digital tech companies, which, unlike digital tech jobs, includes other roles like HR, legal and accounting professions. Together with job advertisement data, we get a more comprehensive view on UK labour markets.’
The longest processes was evaluating the skills. This involved applying weighting to the skills associated with different roles to understand the top skills that employers are demanding. As this is text data, running this analysis for several locations took a while.
The shortest process was calculating the median salary for the roles - this is simply because there is an in-built function in R which is ready to use.
What I find takes a bit of time is thinking about how to visualise the data - because you have to think about the audience and the best way to display the data in such a way that it requires minimum effort from the reader, but at the same time encapsulating all the information you want the reader to absorb.
Lucy Cousins, the data design manager in the insights team, builds the visuals using JavaScript to make them all look really nice and top spec.
This is what Lucy had to say about working on the visualisations.
’When I create visualisations it’s important that they’re clear and can be understood by the user. Visualisation is such an important part of communicating data and a method I believe speaks to everyone, regardless of their background. So it is vital that what I display can be interpreted by anyone, making our insights available to everyone.’
Dominic Darcy, who works in design, Paul Abbot and Anne-Marie who work in the development team and Orla Browne the content editor all play an important roles in the report from the design, web design and build and the content.
The report of the research and analysis was launched at a Tech Nation event on June 13th 2019.
Did you find this content useful?
Thank you for your input
Thank you for your feedback
DataIQ Transform
You may also be interested in
5th Floor, 10 York Road, London SE1 7ND
Registered in England: 9900834
Tel: [+44] 203-829-1112
Copyright © IQ Data Group Limited 2019