After analyzing the results of this study, there are several points for consideration. One of the first conclusions that can be drawn is that the type of humor utilized on social media has a very limited effect on the emotional reactions of a consumer. As seen in figures 1.1 through 1.8, the emotional reactions to all of the four types of humor, regardless of the format that they are presented in, result overwhelmingly in neutral or positive (happy, amused) emotional reactions. Very few of the study participants indicated that their feelings were negative after being exposed to the humorous content. This is consistent with findings presented by Taecharungroj & Nueangjamnong (2015), as they concluded through their research that the use of humor results in positive reactions. The results also indicated that of the different types of humor, Personification faired better than the other three types of humor in both formats when it came to the emotional responses. The positive emotional response levels (Happy, Amused, Neutral) were higher for Personification than any other humor type. While this is inconsistent with the findings of Taecharungroj & Nueangjamnong (2015), which made the claim that ‘Surprise’ humor would be the most popular amongst consumers, Personification was the clear favorite among participants in this study. It may in part have been due to the similarity between the small sample population demographically, however this conclusion remains a valid one. It is possible that Personification was the most well recieved due to it's ability to appeal to many different senses of humor. The example tweets that were utilizing Surprise humor may not have been as universally considered "funny", which may offer an explanation as to why the findings of this study vary from the findings of Taecharungroj & Nueanjamnong (2015). While it is not possible to come to a solid conclusion as to why this subset of the target population recieved Personification humor the best, it is clear that it was favored above the other three types. Thus, researchers should be sure to take advantage of personification humor, as well as surprise humor, within their social media marketing strategy.
Outside of the emotional responses to the use of humor on Twitter, this study also has brought about points of consideration within the realm of digital engagement behaviors. In this study, participants indicated that across all of the four humor types, their propensity to exhibit high levels of engagement is low, while their propensity to exhibit low levels of engagement was high. This is consistent with conclusions drawn by Chu, Chen & Sung ( 2015), who found that there are differing levels of engagement, all of which often do not occur at the same time. Chu, Chen & Sung (2015) found that consumers will engage with a piece of content in many ways, but that their choice of how to engage with the content is entirely independent of what others may choose. While this study found that participants were less likely to engage in more active ways, such as sharing or commenting, with branded humorous content, that did not mean that they would not engage in less active behaviors, such as favoriting. While it cannot be concretely proven, the data suggests that less active modes of engagement were more likely to occur, which could have to do with the fact that it takes less energy and a lower level of consideration. A person that cares or reacts more strongly to a piece of content is more likely to engage with it at a higher level, which suggests that while consumers enjoy humorous content, their reaction is not strong enough to instigate high levels of enagement. Throughout the results, it was shown that while consumers were far more likely to interact with humorous content in less-engaging ways, the majority would engage with tweets that utilized comparison humor even less so. Across the board, Comparison humor showed low engagement levels, as well as the lowest levels of positive emotional reactions. This could suggest that social media marketers should avoid utilizing comparison humor in their strategy, and instead should choose to employ one of the other three popular types of humor.
This study also served to identify a relationship between brand attitude and the use of humor in a Twitter marketing strategy. Across the board, participants indicated that after viewing a humorous tweet, their opinion either did not change (57.4%) or was more positive (38.7%). This is important information, especially for social media marketing professionals. It remains consistent with the findings of Barry & Graça (2018), who found that in a majority of cases, humorous content would outperform non-humorous content in advertising attitudes. While this study did not look at a comparison between the non-humorous and humorous content, the resulting attitudes of participants in the sample population were a majority positive, which is directly in line with the conclusions drawn by Barry & Graça (2018) in their study. As many remain wary of the efficacy of humor on social media platforms, it is important to note that utilizing humor rarely (3.9%) has a negative impact on brand opinion and attitude.
While this study is indicative of the thoughts, feelings, and potential behaviors of the population sampled, due to a small sample size, the results cannot be generalized to a broader audience. As the nature of this study is to look at consumer behavior, as well as social media marketing tactics, further research should focus on gathering a larger, and more diverse sample size. In this research, the majority of participants were of a similar age, (18-24) and gender (majority female), which could have led to a bias in the results.
In the construction of this study, there remains room for improvement. Throughout the survey that was administered, there were eight examples of tweets coming from brands. While these example tweets were useful for this study, there remains something to be desired when it comes to the prevention of potential bias. Each of the example tweets was part of a set of two, which was delegated by the four different types of humor. In order to prevent brand-bias, future researchers interested in this subject should use pairs of tweets from the same brand. This will eliminate the possibility that responses from participants could be swayed by the brand itself.
Another limitation of this research was the scope of the humor that was being studied. While this study looked at the four most popular types of humor used in social media marketing, there are actually seven types of humor in total. For the purposes of this study, it was important to par the types of humor down into only the most common types of humor, further research could expand their study to identify all seven of the types of humor, and their relationship to brand attitude, engagement, and emotion.
While there are many limitations to this study, the internal validity remains high. The types of humor that the study intended to look at were properly represented by the example tweets, whic allowed the study to accurately answer the proposed research questsions. The research instrument measured for the different points of interest, and offered valid results, despite the small and demographically similar sample population. As previously mentioned, this study cannot be generalized to a larger audience, and thus has a lower level of external validity. With a small sample size such as this, it is not possible to conclude that the same experiment with a more diverse or larger sample population would yield the same results.