From ancient Egyptian hieroglyphics to modern Chinese characters, communication through symbolism is something that’s been prevalent throughout our history. In today’s Digital Age, communication symbols have evolved to the point of ubiquity. Articulating emotive vibes can sometimes get tricky, and small digital symbols otherwise known as “emojis” have become a playful way to help us better communicate a message’s tone. Emojis are commonplace in today’s communication via text messages, e-mail, and social media; but should they have a place in research? In survey-based research, it’s essential that all communication is well-defined, thus understanding how emojis impact research outcomes is significant. As such, FocusVision partnered with The Advertising Research Foundation on a study to explore:
1. How the presence of emojis influence…
- survey dropout rate
- length of interview
- survey satisfaction
- data comparability
2. How people perceive various facial emojis
“Picture” Perfect Methodology
To gain an understanding of how emojis impact survey-based research, we interviewed an online sample of 2,000 individuals, representing a cross-section of the United States population. The respondents were recruited through Research Now. We designed a “text-based” survey and an “emoji-based” survey, using Decipher®, FocusVision’s online survey solution. Both surveys had identical content; the only difference being that the “text-based” version had text only, and the “emoji-based” version had both text and emojis.
Our rationalization for including both text and emojis in the “emoji-based” version was grounded in the fact that humans are the only species that can communicate by verbal and written language, a truly extraordinary skill. Accordingly, text was not something we sensed ought to be detached from the research process. Text allows us to convey incredibly specific information that would in most cases be impossible to communicate through emojis alone. For example, in the illustration below “Definitely will eat” and “Probably will eat” have two very different meanings, which can only be communicated by means of text.
Two Thumbs Up
When comparing the two survey types, there was no significant difference in drop-out rates or interview length. We did, however, see slight variations in survey satisfaction. Respondents rated the “emoji-based” version of the survey a tad more enjoyable than the “text-based” version; a difference of only three percent, but still statistically significant. We then probed discrete segments of the population. When exploring age group differences, we found that Millennials and Generation X respondents considered the “emoji-based” survey to be more enjoyable than the “text-based” version. There was no significant difference observed amongst Baby Boomers. When looking at gender differences, we found females viewed the “emoji-based” survey more enjoyable than the “text-based” version. There was no significant difference found amongst males.
That Looks Good Enough to Eat
Perhaps the most noteworthy finding from our study was discovered when we looked at how respondents rated their level of satisfaction on a number of different foods. We tested an assortment of 30 foods, all of which scored higher in satisfaction for the “emoji-based” version versus the “text-based” version. This is significant because one would presume that in order to sway food satisfaction scores, an image of an actual mouthwatering burger, for example, would be a requirement.
As it turns out, a pixilated “burger emoji” was all that was needed to sway food satisfaction measures. We speculate this could be because emoji images may make it easier for participants to visualize foods. Likewise, when respondents were asked how likely they are to eat various foods within the next 30 days, those who saw the “text-based” version were more likely to say they would eat unhealthy foods, than folks who saw the “emoji-based” version. This could be because many people usually want to avoid unhealthy foods, and since it’s harder to visualize foods through text versus emojis, an inverse effect was observed.
The key takeaway here is that participants in our study did indeed respond differently to emojis versus text under certain conditions. Thus, when conducting repeat studies such as trackers, it’s important to be mindful of the fact that images may yield different results. In Part II of this blog series, we’ll investigate how people perceive various facial emojis. The results may make you .