So, I’ve just completed a draft of the research project.

I used the definitions provided by participants to code data as either direct, indirect, or nonexistent – in relation to Barthes’ threefold image. Direct implies that participants acknowledged the literal structure of smileys and explained why people use them; indirect implies that participants explained why people use smileys but said nothing of their literal structures; and nonexistent implies that participants described only the denoted image and failed to explain why people may use smileys.