ASC Conference Review – Tackling Data Overload
We recently attended a conference, ‘Tackling data overload: Making sense of complex multi-source data’ run by the ASC (Association for Survey Computing).
It was a great day with seven expert speakers, from companies including The Times, Network Research and Lightspeed GMI, who specialise in gathering, organising and mining research data or creating visuals from data.
The one day conference addressed the challenges and opportunities of dealing with the current of state research data in the Market Research Industry.
Research data is no longer the typically clean and curated survey data we are used to, it has now grown to a complex combination of pictures, videos, texts and other material.
One of the talks we found the most interesting was, “Big Data is more than behaviour: Why business needs to get smarter about extracting insights from big data” by Colin Strong from Verve.
Colin discusses how almost anything can be put into some form of data. Therefore there is always vast potential and opportunity for Data visualizations. However, he states that big data is rarely used successfully. It can be challenging to pick out the relevant data from the vast amounts available.
These challenges, Colin states, can be tackled by placing greater emphasis on prepping and cleaning the data prior to creating visualizations.
Another particularly useful talk was, “The Icon Experiments primary research exploring the science of visual communication” from Jon Puleston and Satsuki Suzuki, Lightspeed GMI.
Jon and Satsuki discuss why visuals are so useful in conveying information. Visuals are processed much faster than data or text allowing information to be easily digested as we mentioned in a previous blog post. Visuals also hold viewers attention more and makes them more memorable.
They go on to discuss the use of icons. Icons are more effective the longer they have been in the public domain, as the intended meaning will be quicker and easier to identify. Therefore, they suggest, avoiding icons with unique designs.
Conclusions from other talks regarding data visualizations include:
- Organisations should value the importance of an effective data management process. This makes the vital task of cleaning up and prepping data significantly easier and less time consuming.
- Data visualization tools should combine the best of Business Intelligence & MR tools. They should deduct the functionality and practices of both. The aim is to present/visualise data in a way that is fast, robust, accurate whilst also being beautiful.
- There needs to be a proper taxonomy system in place in order to effectively process video data for research.
- Government agencies usually strive on making their data visualizations accessible to the whole public and should be referred to if your intent is similar.
- There are constant and impressive developments within the data visualization world which are already evident on newspaper/journalistic websites such as ‘The Times’. Such developments include:
- Integration of advanced statistical models
- Cutting edge data visualization software
- Effective data collecting & data management processes
- Innovative research methods.
For more information on the conference click here.