The focus of this research paper was to determine if the activeness or passiveness of a social media user has an effect on how well the user remembers and finds interest in an advertisement. To find if or how these variables were related, high school students were randomly selected to participate in a two-part survey. In the first half of the survey, the students self-evaluated how active they were on social media using six questions with answer bubbles that could produce scores that were added together to find a quantifiable “social media score” for each participant. In the second half, students watched a five-to-ten minute video on YouTube that was preceded by a short video advertisement, as one would see if he or she were watching a video on the platform on his or her own. After the full video had played, the students were asked a few questions about their experiences with the video and about advertisements in general. They reported if they remembered the advertisement and how interested they were in it after the full video had played. These answers could then be quantified and compared to the social media scores using the Chi-Square test and two-proportion z-test to find possible connections between the variables. The evidence was not strong enough to conclude that active or passive social media use had an effect on how well the students remembered the advertisements, but it could be concluded that passive users found advertisements more interesting than did more active users. These conclusions mean that an advertiser attempting to appeal to high school students might not find a difference between how active and passive social media users in the audience remember and apply the information in an advertisement, but they might reasonably be able to earn the interest of passive users more effectively than active users.
Introduction and Predictions
This study was an investigation of the possible relation between the activity of users on social networking sites (SNSs) and how these SNS users, specifically high school students, interact with advertisements on the platforms they use. This could be important information for marketers, as modern advertisements must appeal to specific demographics in order to be successful. In recent years, many companies have begun to adopt and maintain an online marketing campaign in order to expand and develop a wider consumer base1. This online consumer base is more complex to manage and appeal to than the traditional audiences that businesses marketed to before the advent of the social internet2. One specific factor related to online behavior is the excessive use of SNSs and how different types of gratification, such as content, social, process, and technological gratification, have an effect on how users select and consume their content online3. Understanding how these elements relate to and influence each other is important in creating advertisements that appeal to more people, especially those who overuse social media.
The investigation specifically looked at the disparity of advertisement success between more frequent and less frequent users of social media. As overuse of social media is caused by the allure of gratification and advertisements must provide some sort of satisfaction to be successful3, there may be a connection between how users find gratification and how they perceive the attempts to appeal to them by advertisers. In this situation, the independent variable is how much time one spends on SNSs and the dependent variable is the appeal of a random advertisement. There are conflicting ideas on how the users will interact with the advertisements, but the hypothesis is that there will be a negative relationship between social media use and advertisement retention because an active user would probably be conditioned to take in a lot of information and tune out the less pertinent grabs for attention.
The field of study is bounded in that the understanding of the science and reasoning behind social media use and overuse is an extensive process, especially in studying how the logic applies to each individual participant. So, the focus was specifically on whether increased use of social media and recognition of advertisements are correlated. From there, further testing would be required to understand which specific types of gratification are most successful. However, this process eventually becomes a matter of specifically targeted advertisements, which is a much more in-depth process than this investigation.
This field of study has a few holes the study hopes to address. One missing topic is the connection and interaction between the social media platforms, advertising strategies, and consumers. While this is not what this paper specifically focuses on, it could be important to investigate if there is a change in the techniques used by advertisers or effectiveness of advertisements across social networking platforms. Another hole in the field is how those who are already deeply engaged with social media interact with SNSs. This is closely related to this topic as the study is not testing how addictive habits develop toward social media, but rather how they affect the use and perception of media.
The purpose of this study was to see how the use of social media affects how a user retains and interacts with advertisements. To test this, a survey was developed for random students at Williamston High School in Williamston, Michigan in order to find connections between self-scored habits of both social media use and advertisement interaction. To test retention of advertisements, an artificial environment which presented a video advertisement followed by a short YouTube video was used to measure if participants could remember general information about the advertisement, such as whether or not they found it interesting, after finishing the video. Then the results were pooled to look for a correlation between the variables of activeness and retention. These tests were repeated with thirty-six total samples to increase consistency in the study’s results and improve accuracy.
The first test used was a survey to find out information about the participants’ use of social media. To measure their social media activity, they were asked to answer questions regarding time spent planning to use social media, having an urge to use social media, using social media to forget about personal problems, attempting to reduce social media use without success, becoming restless or troubled when prohibited from using social media, and social media use having a negative impact on a job/studies. These aspects were measured by taking a cumulative sum of six questions on a scale of 1 to 5, with 5 as the most extensive and 1 as the least extensive. These aspects could be scored and quantified on a cumulative 6-30 scale, with 6 being the most passive score and 30 being the most active.
A second test was then taken in the artificial environment immediately after the first test. In this environment students were shown the YouTube video that was preceded by an advertisement. Following the conclusion of the YouTube video, they were asked two “yes or no” questions and one five-point spectrum question: did the participant remember the advertiser and/or product after the short video was complete? (yes/ no); had the participant heard of this advertiser and/or product before? (yes/ no); was the participant interested in what was being advertised? (not at all/ not really/ neutral/ somewhat/ very much). These questions were intended to show the rate of retention among the students, the significance of brand recognition, and the amount of interest the students showed in the advertisement.
The selected participants took the social media use survey by using simple random samples of Williamston High School students. They were randomly drawn from the list of sixth hour classes at the school and then students were drawn from the selected classes. Each teacher was contacted via email or in person to find out if they were willing to let their students participate and then groups of three students were randomly selected from each class of an agreeing teacher.
Since the questions about the attention paid to advertisements were either answered continuously (on a 1 to 5 scale) and dichotomously (yes or no), two different tests were used to compare them to the participants’ social media scores. For the continuous questions, the data was compared using the Chi-Square significance test, since it was measuring a personal advertisement score against the social media group a participant found himself or herself in4. For the dichotomous questions, a two-proportion z-test was used, as it measured the proportions of how many members of each social media group (active or passive) answered either yes or no to a question.
Assistant Professor of Neurology Dar Meshi at Michigan State University helped provide suggestions for the process in developing survey questions and recommended the development of an artificial environment for participants.
Our independent variable was a measure of social media use on a 5-30 scale. Dependent advertisement scores were tested against this. This was based on the two groups – if they fell above (active) or below (passive) the median score of 14.5. For Figures 1, 2, and 5, when Chi-Square significance testing was conducted, scores of 1-2 and 4-5 were added together into two respective groups. In order to conduct a more proper Chi-Square significance test, both groups needed to be as close to having five participants in it as possible.
Question 4 of the survey asked “How often do you recognize/watch advertisements on social networking sites?” and then proposed five answers ranging from never (score of 1) to every time (score of 5). Using the Chi-Square formula to test for statistical significance against the participants’ social media scores, 0.3965 was the calculated value. Since the p-value is greater than the study fails to reject the null hypothesis. There is not significant evidence that social media usage has an effect on how often students recognize/watch advertisements on social networking sites.
Question 5 of the survey asked “How often do you interact with/click on advertisements?” and then proposed five answers ranging from never (score of 1) to every time (score of 5). Then using the Chi-Square formula to test for statistical significance against the participants’ social media scores, 0.2899 was the calculated value. Since the p-value is greater than the study fails to reject the null hypothesis. There is not significant evidence that social media usage has an effect on how often students interact with/click on advertisements.
Question 6A of the survey asked “Do you remember the ad that played before the video?” and then proposed two answers yes (score of 1) and no (score of 0). Then using the two-proportion z-test to test for statistical significance against the participants’ social media scores, 0.24825 was the calculated value. Since the p-value is greater than the study fails to reject the null hypothesis. There is not significant evidence that social media usage has an effect on how often students remember the ad that played before the video.
Sample 1: Passive social media users
Sample 2: Active social media users
Significance Level: 0.05
Test: 2-Proportion Z-Test
Question 6B of the survey asked “Have you heard of this advertiser/product before?” and then proposed two answers yes (score of 1) and no (score of 0). Then using the two-proportion z-test to test for statistical significance against the participants’ social media scores, 0.15625 was calculated. Since the p-value is greater than the study fails to reject the null hypothesis. There is not significant evidence that social media usage has an effect on whether students had heard of this advertiser/product before.
Sample 1: Passive social media users
Sample 2: Active social media users
Significance Level: 0.05
Test: Chi-square test for independence
Question 6C of the survey asked “How interested were you in the advertisement/product?” and then proposed five answers ranging from never (score of 1) to every time (score of 5). Then using the Chi-Square formula to test for statistical significance against the participants’ social media scores, 0.04581 was calculated. Since the p-value is less than the study rejects the null hypothesis. There is significant evidence that social media usage has an effect on how interested students were in the advertisement/product advertisements.
The hypothesis estimated that there would be a negative relationship between the activity of a social media user and how well he or she retains information from an advertisement, meaning that a more passive user would be more likely to remember information. However, the data collected generally showed little evidence for this claim, except for one area that measured participants’ interest in the advertisement that was shown to them. For most of the tests, however, there was simply too little of variation between the respective advertisement scores of those in the passive group and those in the active group to be able to reject the null hypothesis.
This data relates with Haida’s and Rahim’s idea that the factors that lead to an advertisement’s success are its informational value, how entertaining it is, and its creator’s persistence2. It can conclude that passive users of social media generally find advertisements more interesting compared to active users, meaning that they probably felt a stronger mix of entertainment and information than did those in the active group. This disparity could play a role in the focus of advertisements in the future, perhaps causing advertisers to attempt to appeal more to those who use social media less, or more passively, than the median amount.
Based on the research, the study cannot conclude that active and passive social media users retain advertisements differently from each other. This is apparent because, in a test to see if participants remembered a short advertisement that played before a five-minute video, the difference between the retention rates of the active and passive users was too small to register as significant in a Chi-Square statistical test. However, it can conclude that passive users generally found those advertisements to be more interesting than did more active users. This claim is based on another test performed about the same advertisements on the same participants as in the aforementioned test, wherein passive users claimed to find more interest in the advertisements shown than active users. This could potentially prove to be an incentive for advertisers to appeal to more passive users, as mentioned.
The results were affected by a number of limitations and uncontrollable factors. One such complication was that the study collected a relatively small sample size of thirty-six students. The study could have adjusted the format of the survey-lottery to draw more students from each class. This issue could have resulted in an accidental reduction in the randomness and distribution of the participants, since the net casted was so small. It might also have been useful to survey more than just the students at one school; it would be interesting to see if the findings hold up across the high school population or in different age groups than teenagers. Another obstacle was that the advertisements that played before the videos were selected by random chance, meaning that the advertisement was not selected for the individual users or standardized for a cleaner comparison. Indeed, it would be intriguing to see if the results were to change with the use of personalized or standardized advertisements.
(1) Assaad, Waad, and Jorge Marx Gómez. “Social Network in Marketing (Social Media
Marketing) Opportunities and Risks.” International Journal of Managing Public Sector Information and Communication Technologies 2, no. 1 (September 2011): 13–22. https://www.seokursu.com.tr/social-network-in-marketing.pdf.
(2) Haida, Amilia, and Hardy Loh Rahim. “Social Media Advertising Value: A Study on
Consumer’s Perception.” International Academic Research Journal of Business and Technology 1, no. 1 (2015): 1–8. https://www.researchgate.net/publication/280325676_Social_Media_Advertising_Value_A_Study_on_Consumer’s_Perception.
(3) Balakrishnan, Janarthanan, and Mark D Griffiths. “Social Media Addiction: What Is the Role
of Content in YouTube?” Journal of Behavior Additions 6, no. 3 (September 14, 2017): 364–77. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5700732/.
(4) “Chi-Square Independence Test – What and Why?” SPSS tutorials, n.d.
1.What is your sex?
a) Male b) Female c) Prefer not to answer
2. What is your age?
a) 13 or younger b) 14 c) 15 d) 16 e) 17 f) 18 or older
3. In how many extracurriculars (sports, clubs, school memberships) and/or jobs do you participate?
a) 0 b) 1 c) 2 d) 3 e) 4 f) 5+
4. How often do you recognize/watch advertisements on social networking sites?
a) Never b) Rarely c) Sometimes d) Most times e) Everytime
5. How often do you interact with/click on advertisements?
a) Never b) Rarely c) Sometimes d) Most times e) Everytime
6. WAIT until the tester asks you the following questions before answering (circle one)
a. Yes / No
b. Yes / No
c. 1: Not at all interested
2: Not very interested
3: No feelings/neutral
4: Somewhat interested
5: Extremely interested
About the Authors
Marshall Besko and Nicholas Rubeck are students at WIlliamston High School in Williamston, Michigan. In their 11th grade year, they conducted a research project about social media and advertisements as part of a research class that allowed for students to create their own ideas and the means for testing them.