Blog Post #3

Results

I quantified the multiple choice survey on a point system to come to the conclusion that anyone with 17 or more points has a parasocial relationship present, and the strength of the relationship increases as the point value increases with the maximum number of points being 85. Through quantifying 51 responses, I was able to draw conclusions. Then, in order to determine the difference between the various demographics and the scores that the responders received on their surveys, three statistical tests were completed. The first test that was used was an ANOVA, or analysis of variance test, where I analyzed the difference between the mean scores of the three different age groups, 18-27, 28-42, and 43-57. The second statistical test that I performed was a T-test in order to find whether or not there is statistically significant difference between the survey scores between males and females. The final statistical test that I performed was also a T-test, this time, to determine whether or not there is statistically significant difference between the survey scores between different ethnicities.

Test #1 - Ages

The ANOVA was used to analyze the difference between the mean scores of the three different age groups, 18-27, 28-42, and 43-57. The null hypothesis is that there is no difference in mean point value for various age groups while my alternate hypothesis is that there is a difference. Essentially, I was able to conclude that the 18-27 age group has a significantly higher mean than the 28-42 and 43-57 age groups as the mean score value for the 18-27 age group is 26.576 and the mean score values for the 28-42 and 43-57 age groups is 12.25 and 13.1 respectively. Additionally, looking at this table, we can see that there is a statistically significant difference between the 18-27 and 28-42 age groups as well as the 18-27 and 42-57 age groups. This can be seen by the respective Q values of 6.14 and 5.77 as well as the respective P values of 0.00021 and 0.00049. On the contrary, there is no significant statistical difference between the scores of the respondents in the 28-42 and the 43-57 age groups. This can be shown by the Q value of 0.36 and the P value of 0.964. Under a 5% confidence level, the fact that the P-values for the first two comparisons is less than 0.05 and the P-value for the last comparison is greater than 0.05 demonstrates that there is statistically significant difference between the 18-27 age group compared to the 28-42 and 43-57 age group, and no statistical significant difference between the 28-42 and 43-57 age groups. Thus, we can reject the null hypothesis, showing that the mean point value difference is statistically significant.

Test #2 - Genders

A T-test was used in order to find whether or not there is statistically significant difference between the survey scores between males and females. For this test, my null hypothesis was that there is no statistically significant difference in mean point score value between the various genders, and my alternate hypothesis was that there is a statistically significant difference. Using a 95% confidence interval on the T-test, I found that T = 0.3075 and the two-tailed P-value is 0.7614. Because I used a 95% confidence interval, any P-value above 0.05 demonstrates lack of statistical significance. The P-value in the case was 0.7614, which fails to show statistical significance. Looking at these graphs, we can see that the medians of both genders are very similar in point value. Therefore we fail to reject the null hypothesis, and it can be concluded that there is not a statistically significant difference between the mean point values between males and females.

Test #3 - Races

A T-test, this time, to determine whether or not there is statistically significant difference between the survey scores between different races. For my survey, all of the responders were either Asian or White, and thus, the null hypothesis was that there is no statistically significant difference in mean point score value between the various ethnicities and my alternate hypothesis is that there is a statistically significant difference. Using a 95% confidence interval on the T-test, I found that T = 2.3501 and the two-tailed P-value is 0.0656. The P-value in the case was 0.0656, which is above 0.05, failing to show statistical significance. In these graphs, we can see that although the median seems higher, the differences between sample sizes caused the resulting P-value. Therefore we fail to reject the null hypothesis, and it can be concluded that there is not a statistically significant difference between the mean point values between white and Asian individuals.

Discussion

Thus, to reiterate, it was concluded that individuals aged 18-27, regardless of race or gender, exhibited the strongest parasocial relationships. The Generation Z 18-27 age group is much more likely to create a stronger parasocial relationship than older age ranges. This is specfically shown by the data in which there is no statistically significant difference shown between varying races and genders; however, there is a statistically significant difference between Generation Z in comparison to the older generations. Although this was expected, it was unexpected that there would be no statistical difference between millenials and Generation X. This highlights that teenagers and young adults should be targeted when spreading awareness of parasocial relationships because they tend to exhibit the strongest relationships; thus, they will exhibit the strongest effects, which may be both beneficial and negative. To ensure that teenagers and younga adults do not form an addiction to their electronics when following along with their favorite influencers, introspection and awareness of such relationships is imperative.

Limitations/Implications

Due to a limited network of family, friends, and peers, the majority of the responses were from women, Asians, and those in the 18-27 or 43-57 age range; thus, the survey ultimately had sampling bias, essentially meaning there was a lack of variation in demographics during distribution. Additionally, the number of survey responses received was lower than intended, sitting at 51, as a result of a lack in networking ability, so the conclusions may be slightly skewed. Additionally, there could also be potential self-serving/response bias, meaning that the respondents may have over or under exaggerated their responses based on what they expected to be the “correct answer.” If this research project were to be recreated in the future, it should be done over a longer period of time and with access to a larger network of individuals to gain more survey responses with an increased demographic variety such as expanding to wider age range from younger than 18 to older than 57 and a greater variety of races and genders. This research can also be continued by observing other demographics such as sexuality and location demographics. Additionally, it is important to recognize that although parasocial relationships are most prevalent in our society due to social media, and thus these relationships are abundant among teenagers and influencers, they are attributed to any one-sided relationship; thus, varying versions of such relationships can continue to be explored and observed in all demographics.

Conclusion

With 4.76 billion social media users worldwide, the effects of this research project reach far beyond the scope of just Generation Z in the United States. Additionally, it is important to note that although parasocial relationships are most typically created through media, they can essentially be created with anything. By further expanding on our knowledge of the existence and effects of such relationships and observing ways in which we can spread awareness of such relationships. In doing so, our society can promote healthy introspection of one’s online socialization and media relations to effectively encourage the fulfillment of necessary human interactions in our current post-pandemic era.

Reflection

Initially, I was curious about behavioral patterns in children. I have previously conducted research on the bilingual advantage in children. While reading on relevant topics of research in children, I found parasocial relationships, and while exploring such relationships, I discovered that there is an overall lack in research on adults. This is ultimately what lead me to my research question, and as I furuther observed such relationships, I realized they cover a wider scope than just social media relationships. Thus, in the future, I hope to observe the wider and less prevalanet scope of parasocial relationships in which children create one-sided emotional attachments to inanimate objects, such as toys.

Additionally, there was significant uncertainty in whether I would recieve results. As a high schooler, other than my parents, I do not know a lot of people who are older than 18; thus, I struggled to think of people to send my survey to. Due to this uncertainty, I decided to utilize the method of snowball sampling in which I sent out my survey to peers and family and asked them to send it to those who fit the agre requirements and may be interested in filling the survey out. I had to change my methods as a result of this uncertainty

Page updated

Report abuse