the Relationship of Facebook Usage and Loneliness
Research Proposal (it was NOT conducted)

Abstract
The popularity of Facebook among internet users has increased, yet at the same time people start noticing the negative life influence that Facebook brings to its users. The present study furthers our understanding in how the use of Facebook affects one’s feeling of loneliness among fresh graduates who are experiencing life changing events. We look at the number of Facebook friends users have, the amount of time they spend on Facebook, the number of status posted, the number of post liked by others, and the number of pieces of personal information they share. We expect to see a strong correlation between general Facebook use and users’ sense of loneliness. However, the number of like received from others might not have a direct relationship with users’ loneliness. Result of our study could be applied to the educational field, where both teachers and students could benefit. Introduction As social media is getting more popular, Facebook is one of the few social networking sites that have generated great attention. It provides people a platform for communication and a wide range of functions, such as posting status reflecting their daily lives, liking other’s status or photo by clicking the thumb, and to share users’ personal data with others. According to the Statistics Portal (2014), Facebook is the world's leading social networking websites with 1,280 million users worldwide.

However, a certain amount of people start questioning the influence that Facebook does to their life. It seems that Facebook affects users’ well-being negatively. The New Yorker published an article about how Facebook makes people unhappy, with the support of a research study done by the University of Michigan, in 2013. It pointed out that the well-being of Facebook users isn’t as good as how they present to be, yet they are rather unhappy. This shows that a correlation exists between Facebook usage and negative emotions.

Our research is done based on our interest in the relationship between Facebook use and people’s loneliness, since loneliness is also a negative emotion. Several studies have been conducted to look into the correlation between loneliness and Facebook usage.
To begin with, Ryan and Xenos (2011) conducted a study to investigate the relationship between Big Five, shyness, narcissism, loneliness, and Facebook usage. They found out that Facebook non-users were more likely to experience social loneliness.

Secondly, a study that examined the reciprocal relationship between loneliness and Facebook use among first-year college students was done by Lou and her team in 2012. They tried to find out whether loneliness encouraged students to spend more time on Facebook. It showed that higher Facebook intensity reduced students’ loneliness. (Lou et. al., 2012).
Last but not least, Jin wanted to explore how lonely people use and perceive Facebook in 2013. According to Jin (2013), Facebook users’ use time was not significantly relevant to loneliness.

However, some research gaps have been identified from these studies. Firstly, target groups of these studies were too broad. Participants of both studies done by Ryan, Xenos (2011) and Jin (2013) were aged 18-50. This allowed confounds because every participant had a very different background from each other. The results were not significant enough. As for the research conducted by the team of Lou (2012), the target group was first-year college students. The result could only be generalized to this age group who is experiencing college freshman's life. Second of all, these researches took place in foreign countries (USA, Australia and Korea). Results of these studies could only respectively be generalized to these particular locations.

Therefore, the aim of our study is to examine the correlation between Hong Kong fresh graduates’ feeling of loneliness and Facebook use. It is hypothesized that there would be a strong correlation between those two factors, where the more time they spend on Facebook, the more lonely fresh graduates felt. Data for our project will be collected from participants through questionnaires and handled with correlational method.

Methods
Participants
A sampling frame including names, e-mail address and phone numbers of 500 fresh graduated students (FGS) from graduates of bachelor degrees will be obtained through the Office of Polytechnic University with simple random sampling. With this method, each individual is chosen randomly and entirely by chance, such that each individual has the same probability of being chosen. After a year, the survey attached with the inform-consent form will be distributed to those FGS. The target group will be 22-26 years old young adults, who are facing life changes of switching their role from student to employee. They may find it difficult to locate their identity and career paths, so they might go through frustrations. This study is about how changes in real life affect the virtual one (the pattern of using Facebook) and loneliness.

Design
The study is going to use correlational design to examine the correlation between the predictor variable (x) and criterion variable (y). The predictor variable is the pattern of using Facebook. The criterion variable is the loneliness of participants. Furthermore, the study will distribute surveys to 500 participants for data collection. One of the advantages of conducting surveys include having standardized questions that facilitate data analysis. Also, it increases the possibility of administering the questionnaire remotely and lowers the cost of study.

Measures The predictor variable (x) is defined as x1 = number of friends on Facebook (ratio); x2 = amount of time spent on Facebook a day (ratio); x3 = number of status posted a day on average (ratio); x4 = number of posts liked by others per post on average (ratio); x5 = number of pieces of personal information shared, for instance residential address, mobile phone number and so on (ratio).

As for the criterion variable (y), it is defined as loneliness (interval). Loneliness is defined as one’s desire of social relationships that could not be met, it results in negative emotions or desolated feelings (Lou et. al., 2012).The Russell’s UCLA Loneliness Scale is used to measure subject’s loneliness. The UCLA Loneliness Scale consists of 20 items. Each of the items is a question reflecting on the individual’s experience of loneliness. Subjects have to answer questions with four options: never, rarely, sometimes, and always, regarding how often they have the feeling of either lonely or non-lonely. 11 items are negatively worded and 9 of them are positively worded. It requires reverse coding. Each of the options has their own respective scores. For example, never = 1, rarely = 2, sometimes = 3, and always = 4. As for reversed coding, never = 4, rarely = 3, sometimes = 2, and always = 1. The scores are then summed together. The result indicates that high scorers experience greater feeling of loneliness, while low scorers tend to experience less loneliness.

The UCLA Loneliness Scale was evaluated and was found to be highly reliable (Russell, 1996). The test-retest correlation is 0.73 over a one-year period. The correlation is strong, which shows that the performance of the subjects are consistent over time. As for the internal consistency, Cronbach’s alpha for the items is around 0.89 to 0.94. The coefficient alpha implies that the items in the scale is homogenous and consistent.

In terms of validity, Russell’s scale has met the convergent and construct validity (Russell, 1996). The correlation of scores with the other scales, such as the NYU Loneliness Scale and Differential Loneliness are positively correlated. Negative correlations have also been found between loneliness and the measure of social support. These correlations prove that Russell’s Scale meets the convergent validity. Evidence have also been provided to confirm the construct validity of the current scale. Scores of the Russell’s Scales are found to be significantly related to Eysenck Personality Inventory. Furthermore, strong correlations exist between loneliness and self-esteem and depression.

Procedures We are going e-mail a survey, a cover letter and an informed consent document to each of the 500 FGSs chosen by random sampling. The consent form allows participants to authorize us to collect and use their personal information (study and job details) and emotional changes. The survey consists of three parts - Facebook usages, Loneliness scale, and Demographic information. Two days after the email is sent, we will send a reminder note to non-respondents; four days later, we will e-mail the same document pack to all who still have not responded; six days later, we will again send an email informing the non-respondents that a study investigator would call to conduct the survey by telephone. Telephone survey would be conducted on the seventh to ninth days by senior social science students. To maintain confidentiality, all interviewers are needed to sign the secrecy agreements, and only the surname of interviewees are known in order to maintain participants’ privacy.

The telephone survey is a duplicated version of the email survey. The investigator would read the informed consent once before the interview starts. Callers record responses on the computer system. The participants are identified with numbers instead of their real names. The data collections would be finished within ten days.

Data analysis Since the level of measurement for all the predictor variables are ratio, and the level of measurement for the criterion variable is interval, Pearson Correlation will be used to analyze the data. The assumptions are indicated in the followings. It measures linear correlation and the variables are normally distributed. The usage of Pearson’s Correlation can help to examine the strength and the direction of the linear relationship between x and y. The null hypothesis of this statistical test will be r=0, meaning that there is no significant relationship between x and y. The alternative hypothesis will be r≠0, where a significant relationship between x and y is present. Another statistical test, multiple regression, will be used as there are five predictor variables and one criterion variable. Results from this test could also help us make predictions. The assumption of this study is that there is an existence of a linear relationship between the variables. If there is a significant correlation between the predictor and criterion variable, we could predict the score of a variable after knowing the score of another variable (Goodwin & Goodwin, 2013). Performing a multiple regression analysis could also assist in finding the coefficient of determination. It can tell us how well the five predictors explain the result of loneliness. The null hypothesis will be β1=β2=β3=β4=β5=0, meaning that the five predictor variables cannot significantly predict the criterion variable. The alternate hypothesis will be βj ≠0, implying that at least one predictor variable can predict loneliness. Expected Results
Previous Data Previous research concluded that Facebook users have lower social loneliness but higher family loneliness, while Facebook non-users have higher social loneliness (Ryan and Xenos, 2011). It seems that those users are encouraged to build more relationships with other people apart from family members at the expense of relationship with family (Marche, 2012). Still, it is suggested that using Facebook can somehow make people feel lonelier.

Our estimation
Regardless what previous studies or news are claiming, we expect that there will be a positive relationship between the time spent on Facebook and users’ loneliness.

We have considered the sub-factors of Facebook on loneliness. The sub-factors are the number of Facebook friends, the amount of time spent on using Facebook per day, the mean number of status updated, the mean number of “like” received, and the mean number of post shared, correlating to loneliness.

Based on the sub-factors, it is expected that there is a positive correlation between number of Facebook friends and loneliness. The statement “Facebook friends are not all real ‘friends’” is the general consensus. Although people are friends on Facebook, they may not know each other. In some cases, adding friends on Facebook may be an act to make themselves feel less lonely.

Next, a positive correlation between time spent on Facebook per day and loneliness of the users will be expected. It can be explained that they try to use Facebook as a recreational tool to kill the time that they feel lonely during a day.

The number of status posted and loneliness should also be positively correlated. The possible reason for this is that people who feel may want to seek for more attention from others by posting personal statuses.

As for the number of post liked by others, it is expected to be negatively correlated with the degree of loneliness. As more post being liked by others may allow people to feel like they are being attended to. This may have a certain chance in lowering their loneliness.
Finally, the correlation between pieces of personal information shared and user’s loneliness is expected to be positively correlated. Lonely people tend to expose their personal information on Facebook in order to provide more other channels for their Facebook friends to approach them.

Discussion The hypothesis would be supported by our expected results.

Interpretation of Data
In this study, what we expect is that the more time spent on Facebook, the lonelier a user will feel.

Loneliness and Number of Friends
In real-life situations, most of us always focus on individual success in different kinds of aspects, such as career or personal achievement. To permit the success, some of us usually ignore or even sacrifice the others for our own benefits (DeNicola, 2014). Eventually, those people who are highly concentrated on individual success will become more distant from their friends. This results in the increase of their loneliness.

The next thing to consider is quantity. Most humans are explicitly in will of pursuing more and more. That is to increase the quantity (DeNicola, 2014). As the quantity increases, our attention and values towards a person individually within a group become diminished. Take Facebook as an example, as we add more and more friends on Facebook, Facebook leads us to the feeling that we are being sociable. However, how many of the Facebook friends do us actually know and interact with? The answer is “only a little portion”. Most of them are only virtual friends or just added for the sake of having more “friends”. That is also a reason that we assume that the higher number of Facebook friends may be a possible cause for higher degree of loneliness.

After that, social connecting tools, such as Facebook allow us the ability to edit. To make ourselves having a better image, most people tend to choose the best representations to present themselves before they update their status (DeNicola, 2014). Those representations include some sayings, favorite photos, lyrics, poems, etc. This kind of communication, unlike face-to-face communication, shows only a little of their true selves. Facebook friends may find difficulties in understanding who they are, or even communicating with them. It turns out that users become lonelier.

Loneliness and amount of time spent on Facebook The positive correlation between amount of time spent on Facebook and loneliness could be explained by decreased time spent with friends. Instead of spending time with friends, users spend more time on browsing Facebook, the lack of physical contact with real friends may result in the feeling of loneliness.

We also expect to see a correlation which the more time users spend on Facebook, the more social isolation would be caused. Some previous researchers suggest that people who always read through other people’s posts and photos tend to be not satisfied with their own life (Jimenez, 2013). Meanwhile, as people feel lonelier, they tend to use Facebook for recreational use in order to kill the lonely time of a day. There is a reciprocal effect between the frequency of using Facebook and the degree of loneliness.

Loneliness and number of status posted
Krasnova’s (2013) research supports that when users spend more time on Facebook, as opposed to actively creating contents and engage with it , the more envious they would feel. This results is based on the social comparison theory which states that when we read the statuses written on the Facebook wall, people usually compare themselves with the writers.

Loneliness and number of post liked by others Facebook users tend to seek recognition by getting more likes from others. The negative correlation between the two variables suggest that people with more likes received from a post tend to be less lonely. The more likes the users receive, the more attention they feel they are getting. When lonely Facebook users receive likes from the others, they may think that the others care about them and focus on them. This may reduce their feeling of loneliness.

Loneliness and pieces of personal information shared
Facebook users may share their personal information on their profile, such as email address, mobile phone number, residential address, or date of birth. Al-saggaf’s coming study shows that nearly 98% of the lonely Facebook users tend to disclose more personal information (New York Post, 2014). By providing these contact information, these lonely users hope that Facebook friends could contact them proactively.

Methodological Considerations
Despite the advantages of using mail survey, there are also some drawbacks. For example, participants may not able to answer the questions. Secondly, mail survey usually bare a low response rate. These are the reasons why our suggested research would use telephone interview as a back-up plan.

However, telephone survey is also time consuming and costly. It is easily biased because of the presence of the interviewer and socially desirable response. Interview bias can happen when participants give self-report answers, which they may not give the honest answer to the research.

During the correlation research, there may be third variables that the researchers can hardly detect. For instance, personalities of the subjects will not be examined in this study. Personality may be a factor affecting loneliness. Nayyar & Singh’s (2011) study shows that there is a significant relationship between loneliness and extraversion, neuroticism, anxiety and locus of control.

As different participants have different life experiences, their survey results might become an outlier. An outlier result possibly affects the correlation direction and the strength of correlation.

Implication
Facebook may increase certain kind of loneliness. The professors/ lecturers could give some information to teach students how to use Facebook correctly (e.g. be creative, effective, safe). After lectures, teachers could use Facebook as a media to provide follow-up sessions of clearing conceptual problems of lectures and solving difficulties in assignments. They could even provide some counseling services to students that helps relieve anxiety that is caused stressful school life. Lecturers may sometimes open online Facebook groups and randomly invite a group of students to join the group. Within the group, students can share some opinions on different interesting topic they suggest. This allows the students within the group to know and understand each other, which can help reduce the feeling of loneliness among students.

References
DeNicola, M. (2014, March 14). Is Facebook making us more lonely? this video will change the way you look at social media. Retrieved from http://www.collectiveevolution.com/2014/03/14/are-facebook-other-social-media-websites-making-us-more-lonely/

Goodwin, C. J., & Goodwin, K. A. (2013). Correlational research. Research in psychology:
Methods and design (Seventh ed.,). Hoboken, NJ: Wiley.

Jin, B. (2013). How lonely people use and perceive Facebook. Computers in Human Behavior,
29 2463–2470.

Konnikova, M. (2013, September 10). How Facebook Makes Us Unhappy. Retrieved July 25, 2014, from The New Yorker: http://www.newyorker.com/tech/elements/how-facebook-makes-us-unhappy

Lonely people share too much on Facebook (2014). Retrieved July 29, 2014 from http://nypost.com/2014/05/21/lonely-people-share-too-much-on-facebook/ Lou, L. L., et. al., (2012). An examination of the reciprocal relationship of loneliness and
Facebook use among first-year college students. J. Educational Computing Research, 46
(1), 105-117

Marche, S. (2012, April 2). Is Facebook making us lonely?. Retrieved from http://www.theatlantic.com/magazine/archive/2012/05/is-facebook-making-us-lonely/308930/ Nayyar, S., & Singh, B. (2011). Personality correlates of loneliness. Journal of the Indian
Academy of Applied Psychology,37(1), 163-168.

Russell, D. W. UCLA Loneliness Scale (version 3): reliability, validity, and factor structure.
Journal of Personality Assessment, 66, 20-40.

Ryan, T. L. & Xenos, S. (2011). Who uses Facebook? An investigation into the relationship between the Big Five, shyness, narcissism, loneliness, and Facebook usage. Computers in
Human Behavior, 27 (1), 1658–1664.