Editor’s note: Tiffany Tran is a market researcher at social media technology firm, Infegy, Kansas City, Mo.
As social media data continues to be a highly sought after source to derive insight from, more and more market researchers will be required to report, incorporate and deliver impressive accurate findings.
Those that are new or are unfamiliar with analyzing social media data will more than likely make at least one or both of these mistakes — misrepresenting data and forcing the data.
Misrepresenting data is simply stating something of significance when it really isn’t that significant, while forcing the data means jumping to an assumption without further investigation.
To help avoid these mistakes with social media analysis, here are four things that market researchers can put into practice.
Know the differences between sources and posts. The number of sources shows how many people are uniquely talking about the search topic while the number of posts is showing how often the search topic is coming up. An analyst may mix the two up or acknowledge only one number and not the other.
For example, when looking at eight million posts discussing the Olympics and saying eight million people were talking about the games one week prior is incorrect. However, if one were to say that there were eight million mentions online about the Olympics one week prior to the games instead, that would be deemed correct.
Set an “at least” percentage to hit. Setting an “at least” percentage says that in order for a variable (search query) to be deemed as significant, it must be at least 5 percent of the total sources/mentions. The percentage is developed based on the number of sources or mentions that are being focused on.
Determining a solid sample number with social media data can be difficult sometimes because it remains arbitrary to the analyst. Use your best judgment and ask honestly what a good sample size would be based on the number of sources/posts presented. If there is a sample of 100,000 sources, one may go with 5 percent (at least 500 sources) or 15 percent for a sample of 100 (at least 15 sources), since 5 percent may be too low in the second instance. Setting this tip in place when analyzing social media data will help bring more relevant insights.
Ask, “How did I get this answer?” Take the time to walk through your methodology. Talk through each query and break down why that particular query is built the way it is. You might be pleasantly surprised at how you arrived at your answer or maybe even realize something went wrong. Always record your methods, queries and filters that were applied so that you can always revisit them if you need to.
Understand and take into account channel bias. Recognize that certain social media platforms – and the Internet in general – have user bases that are skewed toward particular opinions. To avoid biased conclusions, make sure the collection of data is coming from a diverse set of channels. If heavily focused on one channel, take note and be aware and transparent about the possible biases. Twitter, for example, will have a different kind of audience than Pinterest.
By implementing these tips into the research process, both analysts and market researchers can prevent making simple mistakes that often happen in social media research. Mistakes can range anywhere from missing the mark at a new client pitch to losing thousands of dollars if clients are given incorrect data from the beginning.