Patterns in the Data

Survey data analysis has been known to generate table after table of output. If you follow the approach of “run everything to find anything” the volume of output increases exponentially. It’s times like these, especially if you are developing a report for senior management consumption, that the astute researcher will fall back and create a visual display of the data. Online survey tools such as Cvent or productivity tools such as MS Excel or PowerPoint, or dedicated data visualization tools such as Tableau offer graphing as a means of enhancing our understanding of the data.

The chart below is an example from a B2B marketing research study. It could easily be adapted to fit the needs of consumer marketing research as well. The question format used to generate the data was a side by side matrix where respondents were asked to rate their opinion of the companies in an industry vertical as well as their willingness to recommend a company. A screening question was used prior to this exercise to assess respondent awareness of the companies. Only those who indicated they had used a company or were aware of, but had not used the company were allowed to provide ratings. This means the sample sizes vary for each company. This has impact on the subsequent statistical analysis.

data visualization

The chart above is a plot of average scores for willingness to recommend and overall opinion. The standard deviation and number of responses are other key data points that can be gleaned from Cvent reports or output from statistical packages.

From the chart you can clearly see that ‘Out and About’ lags behind all other competitors in the market. Marketers for this company need to focus their research efforts on better understanding the dynamics of the market and why customers and prospects rate them poorly. Conversely Acme and Beta Mart are in a virtual tie for perceived best in the market. Excel and statistical packages such as SPSS can be used to assess the significance of observed differences. For example, Acme and Lost Horizon share similar scores for willingness to recommend, but respondents thank more highly of Acme.

Data visualization engages a different thought process than observing survey results through tables only. The use of well-designed graphs should not be limited to executive reporting, but can be an effective tool for uncovering insights within the data.

blog comments powered by Disqus
Crimes in Design Webinar
Subscribe to our Monthly Newsletter