Professional Documents
Culture Documents
Zell Moeller 2017 Areyouhappyformeon Facebook Pre Print
Zell Moeller 2017 Areyouhappyformeon Facebook Pre Print
Augustana University
Author Note
Anne L. Zell, Department of Psychology, Augustana University; Lisa Moeller,
Watertown, SD.
This research was supported by grants from Augustana University’s Civitas program and
Augustana Research and Artist Fund.
Address correspondence concerning this article to Anne Zell, Department of Psychology,
Augustana University, 2001 S. Summit Ave, Sioux Falls, SD 57197. Email: anne.zell@augie.edu
Acknowledgements:
Thank you to Chris Schatsneider for generously sharing his time and statistical expertise.
IMPORTANCE OF LIKES AND COMMENTS 2
in personal interactions has been associated with positive outcomes (Gable & Reis, 2010). We
communication (O’Sullivan & Carr, 2017). We surveyed participants (n = 311) about their
Facebook status updates from the previous two weeks. As hypothesized, participants perceived
as more positive and important and recalled better their status updates that had received more
responses. Receiving more likes and comments on one's status updates was also associated with
reporting greater happiness and self-esteem, greater satisfaction with the responses one's status
updates received, and perceiving one's Facebook community to be more interested in one's good
news. The present findings point to the potential importance of the likes and comments people
1. Introduction
Capitalization refers to the “social sharing of positive events” (Gable & Reis, 2010, p.
198). According to Gable and Reis’s (2010) model of capitalization processes, sharing news of
the positive events in one’s life increases subjective well-being (e.g., Lambert et al., 2013) and
fosters relational intimacy (MacGregor, Fitzsimons, & Holmes, 2013; Otto, Laurenceau, Siegel,
& Belcher, 2015). The model additionally states that the benefits of sharing one’s positive news
with others stem at least partly from the audience’s response, specifically how constructive vs.
destructive and active vs. passive it is (Gable, Reis, Impett, & Asher, 2004). As Gable and Reis
(2010, p. 213) put it, “capitalization attempts create an opportunity for the partner to demonstrate
responsiveness to the self”. Consistent with this model, research has found that sharing positive
produced a greater increase in positive mood (Lambert et al., 2013; Gable et al., 2004) and
interacting with others online using social networking sites is widespread, and the quality of
these online relationships seems potentially important, as feeling supported by one’s Facebook
community has been associated with greater subjective well-being (Grieve et al., 2013; Lee, Noh
& Koo, 2013). People post status updates containing positive news often in hopes of receiving
validation from others as a way of seeking support on Facebook (Blight, Jagiello, & Ruppel,
2015). Status updates posted on one’s Facebook wall are masspersonal communications
IMPORTANCE OF LIKES AND COMMENTS 4
(O’Sullivan & Carr, 2017), as they are made available for any and all in one’s social network to
see rather than being transmitted directly and privately to a specific individual. Would
masspersonal capitalization attempts on Facebook yield the same positive effects as personal
capitalization attempts have been found to produce? Choi and Toma’s (2014) finding that
sharing positive events on Facebook was associated with experiencing greater positive affect
implies that they might. Going a step further, the present study focuses on the effect of the
responses people receive to their Facebook status updates in general. We propose that receiving
more (vs. fewer) “likes” and comments on one’s Facebook status updates is associated with
positive outcomes, detailed below in section 1.1 and subsections. To our knowledge, this is the
first study to apply capitalization theory (Gable et al., 2004; Gable & Reis, 2010) to an online
1.1. Possible Benefits of Receiving Responses (i.e., Likes and Comments) to Facebook
Status Updates
How might sharing news with an enthusiastic audience on Facebook lead to positive
outcomes? Gable and Reis’s (2010) model, drawing upon Langston’s (1994) hypotheses,
proposes that capitalizing increases subjective well-being in part because it makes the personal
event seem more significant and increases its memorability and because it gives the audience an
opportunity to exhibit care and supportiveness. Applying this model to Facebook requires that
we first consider differences between offline and online contexts: Whereas Gable and Reis’s
(2010) model assumed that the capitalization attempt would be directed toward an individual, in
the case of Facebook status updates the target of communication is a group. Thus, the present
people who wished to respond to a status update could post comments or employ a one-click
communication in the form of the thumbs-up “like” button. Facebook displays the total number
of comments and the total number of likes that each status update has received. With those
differences in mind, we generated specific hypotheses about how the capitalization processes
1.1.1. Perceptions of one’s own news. The capitalization model states that telling a
supportive person about a positive personal event causes one to value that event more highly and
perceive it as more significant (Gable & Reis, 2010; Langston, 1994). Consistent with the model,
Reis et al. (2010) found that sharing news of a positive event with an enthusiastically responsive
(vs. passive) person face-to-face caused people to view their personal event even more
positively, perhaps because the audience’s excitement confirmed the event’s significance.
Generalizing this to an online context, we speculated that participants would value their
Facebook status update more highly the more responses that particular update had received.
Thus, we hypothesized that receiving more responses to a status update would be associated with
1.1.2. Recall of one’s own news. The capitalization model also states that, because
capitalization conversations typically entail rehearsing and vividly describing the event, they
increase the memorability of that shared event. The increased memorability, in turn, helps to
explain capitalization’s positive effect on subjective well-being: Positive events may have a
greater impact if they are remembered than if they are forgotten. Supporting this line of the
capitalization processes model, Gable et al. (2004) found that people were more likely to
remember a positive personal event later the more people they had told about it, even while
We reasoned that this enhanced memory for more frequently shared positive events could
also occur on Facebook where receiving more responses on a status update might be partially
analogous to telling more people face-to-face. The feature on Facebook that gives “notifications”
each time someone responds to one’s status updates may additionally prompt people to look at or
think about their posts again. Therefore we hypothesized that participants will be more likely to
remember their own status update the more responses it had received. Following the practice in
prior capitalization research, we planned to control for the rated importance of the update in
broadly to how positively people evaluate and feel about their lives (e.g., Diener et al., 2017). As
predicted by the capitalization processes model, capitalization has been found to be associated
with higher subjective well-being (measured by positive affect and life satisfaction), particularly
when the target responded an in active-constructive way (Gable et al., 2004; Gable et al., 2008).
We hypothesized that on Facebook as well, people will experience higher subjective well-being
when the target of their communication (which in this case would be their group of Facebook
friends) is responsive.
In the present study, we were also interested in a construct closely tied to subjective well-
higher self-esteem (Leary & Baumeister, 2000; Reitz, Motti-Stefanidi, & Asendorpf, 2016). On
Facebook, information about one’s popularity is available via the open display of the number of
respondents on status updates. In prior research, perceiving that many people are reading one’s
status updates (Manago, Taylor, & Greenfield, 2012) and the number of responses received to
one’s last three status updates (marginally; Greitemeyer, Mugge, & Bollermann, 2014) have
IMPORTANCE OF LIKES AND COMMENTS 7
been linked with higher self-esteem. Being assigned to receive no response to one’s status
updates (Tobin, Vanman, Verreynne, & Saeri, 2015) or to view the profile of someone else who
had received a great deal of response (Vogel, Rose, Robert, & Eckles, 2014) have been found to
lower self-esteem. On this basis, we predicted that receiving more responses to one’s status
processes model (Reis & Gable, 2010), enthusiastic responses to face-to-face capitalization
attempts have been found to increase intimacy (Otto et al., 2015). Do similar processes operate
on Facebook? The small amount of existing research examining whether receiving responses to
status updates is linked to positive interpersonal outcomes has thus far yielded inconsistent
results: On one hand, Utz (2015) found that the likes and comments people received on their
Facebook status updates did not predict feeling connected to those reading their status updates.
Große Deters and Mehl (2012) found that, although being assigned to post more on Facebook
reduced loneliness, this reduction in loneliness was not attributable to the proportion of status
updates that had received a response. On the other hand, Burke and Kraut (2014) found that
receiving comments, messages, and wall posts from someone predicted increases in feelings of
closeness to that person. Furthermore, Stavrositu and Sundar (2012) found that receiving more
(vs. fewer) comments led bloggers to report a stronger sense of community. Our hypothesis was
that likes and comments—the more the better!—received on one’s status updates convey the
message that other people are interested and are happy that one has positive news. Therefore,
Gable et al. (2004) to measure perceptions that one’s close friend or romantic partner typically
responds with excitement rather than passivity, disinterest, or negativity to one’s news about
positive personal events. It assesses perceptions of responses that vary along two dimensions:
interest and excitement. Passive-Constructive responses are also positive but lack enthusiasm.
responses criticize or undercut the good news. Gable et al. (2004) subtracted the mean of the
Active-Constructive subscale score, and they found this resulting composite PRCA score to be
Further validating the PRCA, Gable, Gonzaga, and Strachman (2006) report that
participants who responded to their partner’s good news in a highly active and constructive way,
as rated by coders viewing videotapes of the interaction, were also rated more highly on the
PRCA by their partners. Thus, they found that objectively observed behaviors such as
“elaboration of positives, linking to other positive events, smiling, laughing” (p. 908), gestures,
We wanted to use the PRCA in a novel way, to measure the perceived responsiveness of
of people rather than an individual, and online rather than face-to-face. We hypothesized that
objectively observable behaviors—in this case, likes and comments on status updates—would be
positively associated with perceptions that one’s Facebook community is typically interested in
and excited about one’s good news, as measured by a modified version of the PRCA. In turn, we
IMPORTANCE OF LIKES AND COMMENTS 9
expected perceptions that one’s Facebook community is interested in and excited about one’s
good news to be associated with greater subjective well-being, consistent with prior research on
1.1.5. Response types. Which are more important, comments or likes? In face-to-face
interactions, positive responses to capitalization attempts have been found to be more beneficial
if they are active than if they are passive (e.g., Reis et al., 2010). On Facebook, writing
comments takes more activity and effort than “clicking like” does. So if Facebook interactions
are similar to face-to-face interactions, we might expect comments, not likes, to correlate with
positive outcomes. Comments (compared to likes) might also prompt more elaboration, which
Gable et al. (2004) theorized is partly responsible for capitalization’s positive effects. Consistent
with this line of thinking, Burke and Kraut (2014, 2016) found that only “composed”
potentially significant ways. When sharing good news with someone face-to-face, one can
usually tell if that person has received the message, and if so, a response is socially expected. By
contrast, when sharing one’s good news on Facebook, lack of response from a particular
individual could stem from that person having not seen the post, or having disapproved of the
post, or having approved of the post but preferring to “lurk” rather than be active on Facebook.
Because people can more easily “get away with” not responding on Facebook than they can face-
to-face, any response at all on Facebook represents more of a choice and therefore carries more
meaning. By this reasoning, we would expect likes, despite requiring minimal effort, to be
associated with positive outcomes. Furthermore, the number of responses (which typically are
IMPORTANCE OF LIKES AND COMMENTS 10
predominantly likes) a person is able to garner on their post may be perceived as being an
indicator of their popularity or of the amount of social support offered by their community. Thus,
we planned to explore whether comments will be more strongly related than likes to the positive
2. Method
2.1. Participants
emailed the survey link to students, faculty, and staff at a small college in the Midwestern U.S.,
and posted it on Facebook and on websites listing online psychology studies. Participants were
offered the opportunity to enter a raffle for a $50 gift card. The survey was begun by 373 and
completed by 311 participants, 77.4% female, 63.5% affiliated with our campus community,
After participants gave informed consent, the opening instructions said, “The first set of
questions will be based on memory. **Please do NOT look at your Facebook account at this
time.**.” Participants then reported gender, age, and whether they had any affiliation with the
authors’ college. They also completed the NPI-13 (Gentile et al., 2013), but it will not be
discussed further because it is not relevant to the present hypotheses and because controlling for
it does not affect any of the results presented here (see Zell & Moeller, 2017, for additional
On the next page, participants read “IMPORTANT: You must NOT be on Facebook at
this time.” Then they were instructed to list all of the status updates they could remember
posting on their own Facebook page in the last two weeks and briefly described the subject of
IMPORTANCE OF LIKES AND COMMENTS 11
each post. The instructions right before they began typing their response repeated, “Please rely
On the next page, participants were instructed to log in to their Facebook account before
advancing to the next page. On the following page, participants reported their number of
Facebook friends, M = 611.24, SD = 424.38, median = 549, range = 12–2711. Then participants
were asked to look at their Facebook page and list all their own status updates from present to
two weeks ago. Twenty-seven percent of participants reported having posted no status updates
within the prior two weeks. We set the maximum number of status updates that could be reported
arbitrarily at 24; a maximum was needed because participants had to scroll past all the blank
fields in order to advance to the next page. The mean number of status updates participants
reported having posted in the last two weeks was 3.41, SD = 4.84, median = 7. A total of 1035
Facebook status updates during the prior two weeks were reported (also reported but excluded
from analyses were an additional 211 status updates that were posted longer than two weeks ago
and 64 reports of Facebook activities other than posting status updates on one’s Facebook wall).
For each status update, participants reported the same information. First, they briefly
described the post. Later, two coders independently rated each post as having been completely
forgotten (0), slightly recalled (1) or fully recalled (2) by comparing it to the list of posts the
participant had reported from memory; M = 0.96, SD = 0.94. Agreement between the coders was
adequate (kappa = .77), and discrepancies were resolved by the first author in consultation with a
third rater.
Next, participants responded to “How positive do you consider the subject of this post?”
with a number from 1 (very negative) to 10 (very positive), M = 8.52, SD = 2.15. They
responded to “How important do you consider the subject of this post?” with a number from 1
IMPORTANCE OF LIKES AND COMMENTS 12
(not at all important) to 10 (very important), M = 6.24, SD = 3.05. They reported the date of the
post. We compared the reported date of each status update to the date the survey was submitted
to calculate how many days earlier the status update was posted.
For each reported status update, participants reported how many likes it received (M =
12.87, SD = 20.11, median = 5, range = 0–175), how many comments it received that they
considered to be positive (M = 2.12, SD = 5.55, median = 0, range = 0–67), how many comments
it received that they considered to be neutral or negative (M = 0.23, SD = 1.12, median = 0, range
= 0–18), and how many comments they made on their own status update (M = 0.59, SD = 2.17,
median = 0, range = 0–32). Only 10% of` status updates were reported not to have received at
least one like or comment. Finally, participants rated how satisfied they were with the response
they received on that post with a number from 1 (very unsatisfied) to 10 (very satisfied), M =
7.26, SD = 2.51.
Several questions pertaining to a different set of hypotheses followed (for details, see Zell
& Moeller, 2017). Then, to measure subjective well-being with a focus on self-esteem and
happiness, we asked participants to rate how much they agree right now with items from the
Rosenberg (1965) trait self-esteem scale (“On the whole, I am satisfied with myself”, “I feel that
I am a person of worth, at least on an equal plane with others”, “I have high self-esteem”, and
“At times I think that I am no good at all” [reverse scored]) and modified from Lyubomirsky and
Lepper’s (1999) Subjective Happiness Scale (“In general, I consider myself a very happy
person”, “I am generally not very happy” [reverse scored], and “I enjoy life regardless what is
going on, getting the most out of everything”), α = .90, M = 5.42, SD = 1.19. Our decision to put
only these seven subjective well-being items on the survey was motivated by concerns about
et al., 2004), with the simplified instruction, “Please rate how accurately each of the statements
below describe your Facebook experience” and the stem modified to refer to Facebook (“When I
post on Facebook about something good that has happened to me…”). The scale contains four
subscales, scored by averaging agreement with the items: Passive-Constructive (e.g., “My
Facebook friends say little, but I know they are happy for me”), α = .68, M = 3.90, SD = 1.18;
friends often seem disinterested”), α = .79, M = 2.89, SD = 1.25; and Active-Destructive (e.g.,
“My Facebook friends point out the potential down sides of the good event”), α = .73, M = 1.99,
SD = 1.06. Following Gable et al. (2004) we also computed a composite score by subtracting
from the Active-Constructive subscale the mean of the other three subscales, M = 1.19, SD =
1.33.
3. Results
We used linear mixed modeling (LMM) to test hypotheses focusing on status update-
level data because status updates varied in number and were nested within participants. We first
conducted a series of linear mixed models (LMM) separately testing number of likes and number
Considered on its own, number of likes received predicted perceiving the subject of one’s status
update to be more positive (estimate = 0.012, SE = 0.003, t(1001.16) = 3.73, p < .001, 95% CI
[0.006, 0.019]) and important (estimate = 0.014, SE = 0.005, t(1003.73) = 2.95, p = .003, 95% CI
[0.005, 0.023]). Number of positive comments received, considered on its own, did not
IMPORTANCE OF LIKES AND COMMENTS 14
significantly predict perceiving the subject of one’s status update to be positive (estimate =
0.010, SE = 0.012, t(1019.23) = 0.87, p = .39, 95% CI [-0.012, 0.033] or important (estimate =
0.031, SE = 0.017, t(1016.70) = 1.86, p = .06, 95% CI [-0.002, 0.063]). We then confirmed (see
Table 1) that likes remains a significant unique predictor of status update positivity and
importance even when accounting for other factors. Thus, participants’ perception of the
positivity and importance of one of their status updates was predicted by the number of likes that
status update had received but not by the number of comments it had received.
Using LMM, we tested separately number of likes and number of positive comments as
predictors of remembering status updates. Ability to recall a status update later on was predicted
by the number of likes it received (estimate = 0.007, SE = 0.001, t(968.81) = 4.65, p < .001, 95%
CI [0.004, 0.010]) and by the number of positive comments it received (estimate = 0.018, SE =
0.005, t(1013.31) = 3.36, p = .001, 95% CI [0.008, 0.029]). However, when we considered likes
and positive comments together in a model that also included control variables (see Table 2),
only likes continued to uniquely predict status memorability; positive comments did not. Thus,
participants’ ability to recall a particular status update was uniquely predicted by the number of
likes that status update had received, but was not uniquely predicted by the number of comments
it had received.
satisfaction with the responses to one’s status updates. Satisfaction with the response received to
a status update was significantly predicted by number of likes it received (Estimate = 0.036, SE =
.003, t(1034.19) = 10.69, p < .001, 95% CI [0.030, 0.043]) and by number of positive comments
IMPORTANCE OF LIKES AND COMMENTS 15
it received (Estimate = 0.066, SE = .012, t(924.57) = 5.63, p < .001, 95% CI [.043, .088]) when
each predictor was examined separately. When likes and positive comments were considered
simultaneously while also controlling for other variables, likes remained a significant unique
predictor of satisfaction but positive comments did not, as shown in Table 2. Thus, participants’
satisfaction with the responses to each of their status updates depended upon the amount of likes
those status updates received, and more so than it depended on the amount of positive comments
they received.
We had hypothesized that reporting receiving more likes and positive comments would
be associated with reporting higher subjective well-being. Subjective well-being was correlated
with mean number of likes received (r = .14, p = .03) and with mean number of positive
comments received (r = .17, p = .01). On an exploratory basis, we computed a new variable, the
total number of likes and positive comments each participant received on his/her most popular
post. We found that subjective well-being also correlated with the total number of likes and
comments received on the person’s most popular status update (r = .21, p = .002).
Using regression, we examined whether the relationship between the response received to
one’s Facebook status updates and subjective well-being would remain when we also accounted
for other factors. As shown in Table 3, the total number of likes and positive comments on the
participant’s most popular post was a significant unique predictor of subjective well-being even
when accounting for gender, age, number of Facebook friends, and frequency of posting on
Facebook. We found a similar but weaker pattern of results when we replaced total number of
likes and comments on the person’s most popular status update with mean number of likes
received (β = .14, t(216) = 2.12, p = .04) or with mean number of positive comments received (β
IMPORTANCE OF LIKES AND COMMENTS 16
= .14, t(213) = 2.04, p = .04). Thus, the more likes and positive comments participants reported
having received on their recent status updates (and particularly on their most popular one), the
Before conducting further analyses with the PRCA-F, we first factor analyzed its items to
check that they loaded onto four subscales as expected. Principal component analysis with
varimax rotation did yield four factors: Passive-Destructive (Eigenvalue = 3.25, 27.08% of the
(Eigenvalue = 1.45, 12.08% of the variance) and Active-Constructive (Eigenvalue = 1.05, 8.77%
of the variance).
The PRCA-F composite score was not significantly related to mean likes (r = .10, p =
.15) or mean positive comments (r = .11, p = .12), but was correlated with the total likes and
comments on the person’s most popular post (r = .16, p = .02). Using regression (see Table 3),
we confirmed that the association between the total likes and positive comments on participants’
most popular status updates and their PRCA-F composites scores remained when accounting for
other factors.
Because the PRCA-F composite score did not correlate as strongly and consistently as we
had expected with the responses received, we carried out exploratory analyses looking at each
PRCA-F subscale separately. The Active-Constructive subscale was not significantly correlated
with mean likes (r = .10, p = .14), mean positive comments (r = .08, p = .27) or total likes and
comments on the participant’s most popular post (r = .11, p = .11) as we had assumed it would
be. The Passive-Constructive subscale showed no significant correlations (ps > .10), as we
expected. The Active-Destructive subscale, to our surprise, correlated positively with mean likes
IMPORTANCE OF LIKES AND COMMENTS 17
(r = .13, p = .05) and mean positive comments (r = .17, p = .002), but not with mean neutral or
negatively with mean positive comments received (r = -.21, p = .002) and with the total number
of likes and comments received on one’s most popular status update (r = -.21, p = .002), though
not significantly with the mean number of likes received (r = -.11, p = .12). As shown in Table 3,
we confirmed that the association between total likes and positive comments on participants’
most popular posts and their scores on the Passive-Destructive subscale remained when we
account for other factors. Replacing this variable with the other measures of response to one’s
status updates, we found that mean number of positive comments received also uniquely
predicted scores on the Passive-Destructive subscale (β = -.17, t(213) = -2.52, p = .01) but mean
Thus, participants’ perception that their Facebook friends are interested in and care about
their good news (as measured by low scores on the Passive-Destructive subscale of the PRCA-F)
was predicted by the average number of comments their status updates had been receiving and
also by the amount of likes and comments received on their most popular status update, but not
by the average number of likes received. The PRCA-F composite scale score showed a more
modest relationship with the amount of likes and comments received on participants most
popular status update, and was not predicted by mean likes or mean comments. The weaker
associations with the composite score appear to be due to the fact that the Active-Constructive
and Active-Destructive subscales did not relate to likes and comments received in the way we
had expected.
IMPORTANCE OF LIKES AND COMMENTS 18
3.5.1. PRCA-F and satisfaction with responses. Mean satisfaction with the responses
received to one’s status updates correlated with the composite PRCA-F score (r = .23, p < .001)
and with the Passive-Destructive subscale (Passive-Destructive r = -.27, p < .001; Active-
p = .08).
3.5.2. PRCA-F and subjective well-being. Subjective well-being correlated with the
composite PRCA-F score (r = .28, p < .001) and particularly with the Passive-Destructive
subscale (Passive-Destructive r = -.36, p < .001; Active-Destructive r = -.24, p < .001; Passive-
regressions accounting for participant gender, age, number of Facebook friends, and number of
posts in the last 2 weeks revealed the same pattern, with subjective well-being significantly
predicted by the composite PRCA-F score (β = .27, t(301) = 4.71, p < .001 ) and the Passive-
Destructive (β = -.35, t(301) = -6.20, p < .001), Active-Destructive (β = -.23, t(301) = -4.06, p
< .001), and Active-Constructive (β = .12, t(301) = 2.09, p = .04), but not Passive-Constructive
(β = .01, t(301) = .17, p = .87) subscales. Thus, participants who reported viewing their
Facebook friends as being interested, supportive, and excited for their good news also reported
4. Discussion
Results supported our primary hypotheses. First, we found that receiving more responses
to one’s status update was associated with perceiving that status update as more positive and
more important. Second, participants exhibited better recall of their status updates that had
received more (vs. fewer) responses. Third, the more responses participants reported having
IMPORTANCE OF LIKES AND COMMENTS 19
received on their status updates, the greater happiness and self-esteem they subsequently
reported. Fourth, the more responses participants reporting having received, the more they
perceived their Facebook friends to be interested in and care about their good news. As we
hypothesized, then, receiving more responses to one’s Facebook status updates was associated
4.2. Is receiving comments more strongly related than receiving likes to positive outcomes?
The capitalization processes model asserts that the positive effects of capitalization
attempts depend at least somewhat on the perceived responsiveness (Reis et al., 2004) of the
partner, in other words, the extent to which the person perceives their partner as paying attention
to them, understanding them, supporting them, and feeling pleased for them. This model was
developed with interpersonal communication in mind. Applying this model to Facebook raises
the question, what kind of reaction to a status update–an online masspersonal communication
Clicking like takes so little effort that one might doubt that likes are valued by their
recipient or have any impact. Consistent with that idea, Carr, Wohn, and Hayes (2016) had found
that the more automatic and mindless likes were perceived to be, the less supportive they were
perceived to be. Burke and Kraut (2014) also found that participants reported being closer to
individuals the more messages, posts, and comments they had received from those individuals,
but that likes received from those individuals did not correlate with changes in closeness.
Aligning with those findings, in the present study it was the number of comments received (not
the number of likes received) that correlated with thinking that one’s Facebook community cares
about and is interested in one’s good news. The greater effortfulness of comments might lead
IMPORTANCE OF LIKES AND COMMENTS 20
them (relative to likes) to be perceived as more meaningful indicators of genuine care and
interest.
However, our other findings suggest that we should not be too quick to dismiss likes as
unimportant. We found that number of likes was actually a better predictor than number of
comments was of perceiving a status update as positive and important, remembering it, and being
satisfied with the response received to it. Great numbers of likes may convey a signal of social
proof affirming the positivity and importance of one’s post. Additionally, the receipt of many
likes on a status update may repeatedly call the author’s attention back to it.
We also found that the amount of response (likes and positive comments) participants
received on their most popular posts was a particularly strong correlate of subjective well-being
and of perceiving one’s Facebook community to be interested in one’s good news. This
exploratory finding implies that people’s global impression that they are well-liked and cared
about by their Facebook community may be more closely related to the greatest amount of
response they received than to the average amount of response that they received, consistent with
peak-end theory (e.g., Fredrickson, 2000). In light of these intriguing but exploratory findings,
we encourage any future researchers in this area to look at the various ways of measuring
responsiveness on Facebook.
4.3. Assessing the modified version of the Perceived Responses to Capitalization Attempts
scale (PRCA-F)
The Perceived Responses to Capitalization Scale was designed by Gable et al. (2004) to
measure perceptions of a partner’s tendency to respond with enthusiasm rather than passivity,
disinterest, or negativity when one shares positive personal events. In the present study we
responds when they share their good news. The present study provides initial evidence for the
validity of this version modified for Facebook (PRCA-F). First, we observed a four-factor
structure that matched the subscales of the original PRCA (Gable et al.., 2004). The subscales
showed fairly good reliability (with the exception of the Active-Constructive, α = .51). Second,
we found that the composite score correlated with subjective well-being and that the Passive-
Destructive subscale showed a particularly strong (negative) relationship with subjective well-
being, consistent with findings obtained with the original PRCA (Demir & Davidson, 2013;
Gable et al., 2004). Third, we found that the PRCA-F composite score correlated with
objectively observable behavior: the total likes and positive comments given on participants’
most popular status updates. Fourth, the PRCA-F composite score correlated with participants’
That being said, we recommend that any future users of the PRCA-F look at the
individual PRCA-F subscales, rather than only at the composite score. In the present study, the
inattentive, disinterested, and uncaring) was the subscale most strongly related to subjective
well-being, amount of responses received, and satisfaction with responses received. Participants
seemed to make inferences about how interested and caring their Facebook community is based
on the amount of responses (especially comments) they received on their status updates. The
other three subscales, however, showed diverging patterns of relationships with our outcomes
variables. For researchers who might be interested in using the PRCA-F in future work, we
Receiving more responses to one’s Facebook status updates was linked with valuing and
remembering what one had posted about, reporting higher happiness and self-esteem, and
perceiving one’s Facebook network to be a more interested and caring community. These results
communication. This suggests that Gable and Reis’s (2010) model of capitalization processes
one redefines the audience as a group and redefines an active-constructive response as a higher
Although the present study was specifically designed to investigate whether the
capitalization processes model (Gable & Reis, 2010) could be applied to an online context, the
present findings may also accord well with other theories. For example, the Motivation
Technology model (Sundar, Jia, Waddell, and Huang, 2015) suggests that to the extent that
technologies afford people opportunities to meet their needs for competence, relatedness, and
autonomy (drawing upon Self-Determination Theory – Ryan & Deci, 2000), people will feel
intrinsically motivated to use the technology. This model states that to the extent that technology
affords opportunity for interaction, people will feel a sense of connection and relatedness.
Consonant with that, we found that receiving more likes and comments from others was
associated with feeling more satisfied with the response received and with perceiving one’s
4.5. Limitations
fact that we relied upon our participants’ honesty and accuracy in reporting numbers of likes and
comments and in reporting their status updates from memory without looking at Facebook.
IMPORTANCE OF LIKES AND COMMENTS 23
Additionally, participants answered questions about how many people liked and
commented on their status updates before answering questions about their subjective well-being.
This may have influenced participants to base their sense of well-being on the number of people
who responded to their Facebook posts more than they normally do in their day-to-day lives. On
the other hand, the fact that Facebook visibly lists the number of people who clicked like or
It is important to note that the present study used a correlational design and cannot
demonstrate causation. Our findings are consistent with the idea that the amount of response
received to a status update affects how positive and important it is perceived to be and how well
confounding variables of participant age, gender, number of Facebook friends, and frequency of
posting, as well as status update age, positivity, and importance. However, as we did not use an
experimental design, we cannot entirely rule out other explanations for these associations. For
example, the present findings are consistent with our idea that seeing that one’s status update
receive many likes enhances the seeming positivity and importance of one’s post. However,
status updates about more (vs. less) important and positive events also seem likely to accrue
more likes and comments, and we speculate that the causal influence runs both ways. Similarly,
although we found that amount of response received correlated with subjective well-being and
with perceiving one’s Facebook community to be more interested and caring, we cannot
determine whether these are causal relationships. Possible confounds include extraversion
(Lonnqvist & große Deters, 2016) and differences in perceived offline support (Trepte, Dienlin,
4.6. Conclusion
IMPORTANCE OF LIKES AND COMMENTS 24
Prior research (Gable et al., 2004; Lambert et al., 2013; Otto et al., 2015) has found that
interactions leads to positive outcomes. This study supports and extends that work by showing
that receiving likes and comments from one’s Facebook community is also associated with
positive intrapersonal and interpersonal outcomes. On a practical note, the present findings,
though correlational, are consistent with the idea that clicking like might help to affirm the
positivity and importance of someone’s news and increase their likelihood of later remembering
that event. Or, that typing a comment on someone’s status update may help to allay the
recipient’s fears that their Facebook community is not actually interested in reading about their
good news. When people share personal news, it is important to them that their audience display
interest and enthusiasm: This seems to be the case on Facebook walls just as it is in face-to-face
interactions.
IMPORTANCE OF LIKES AND COMMENTS 25
References
Blight, M. G., Jagiello, K., & Ruppel, E. K. (2015). ‘‘Same stuff different day:’’ A mixed-
method study of support seeking on Facebook. Computers in Human Behavior 53, 366–
373. http://dx.doi.org/10.1016/j.chb.2015.07.029
Burke, M., & Kraut, R. E. (2014). Growing closer on Facebook: Changes in tie strength through
social network site use. In Proceedings of the SIGCHI conference on human factors in
Burke, M., & Kraut, R. E. (2016). The relationship between Facebook use and well-being
Carr, C. T., Wohn, D. Y., & Hayes, R. A. (2016). [Like image] as social support: Relational
http://dx.doi.org/10.1016/j.chb.2016.03.087
Choi, M., Toma, C. L. (2014). Social sharing through interpersonal media: Patterns and effects
http://proxy.augie.edu:2264/10.1016/j.chb.2014.04.026
Demir, M., & Davidson, I. (2013). Toward a better understanding of the relationship between
012-9341-7
IMPORTANCE OF LIKES AND COMMENTS 26
Diener, E., Heintzelman, S. J., Kushlev, K., Tay, L., Wirtz, D., Lutes, L. D., & Oishi, S. (2017).
Findings all psychologists should know from the new science on subjective well-being.
Gable, S. L., Gonzaga, G. C., & Strachman, A. (2006). Will you be there for me when things go
Gable, S. L., & Reis, H. R. (2010). Good news! Capitalizing on positive events in an
doi:10.1016/S0065-2601(10)42004-3
Gable, S. L., Reis, H. T., Impett, E. A., & Asher, E. R. (2004). What do you do when things go
right? The intrapersonal and interpersonal benefits of sharing positive events. Journal of
Gentile, B., Miller, J. D., Hoffman, B. J., Reidy, D. E., Zeichner, A., & Campbell, W. K.. (2013).
Greitemeyer, T., Mugge, D. O., & Bollermann, I. (2014). Having responsive Facebook friends
affects the satisfaction of psychological needs more than having many Facebook friends.
Grieve, R., Indian, M., Witteveen, K., Tolan, A. G., & Marrington, J. (2013). Face-to-face or
29, 604-609.
IMPORTANCE OF LIKES AND COMMENTS 27
große Deters, F., & Mehl, M. R. (2012). Does posting Facebook status updates increase or
Lambert, N. M., Gwinn, A. M., Baumeister, R. F., Strachman, A., Washburn, I. J., Gable, S. L.,
& Fincham, F. D. (2013). A boost of positive affect: The perks of sharing positive
http://proxy.augie.edu:2264/10.1177/0265407512449400
Langston, C. A. (1994). Capitalizing on and coping with daily-life events: Expressive responses
Leary, M, R., & Baumeister, R. F. (2000). The nature and function of self-esteem: Sociometer
2601(00)80003-9
Lee, K-T., Noh, M-J., & Koo, D-M. (2013). Lonely people are no longer lonely on social
doi:10.1089/cyber.2012.0553
Lonnqvist, J. E., & große Deters, F. (2016). Facebook friends, subjective well-being, social
http://dx.doi.org/10.1016/j.chb.2015.09.002
http://www.jstor.org/stable/27522363
IMPORTANCE OF LIKES AND COMMENTS 28
MacGregor, J. C. D., Fitzsimons, G. M., & Holmes, J. G. (2013). Perceiving low self-esteem in
Manago, A. M., Taylor, T., & Greenfield, P. M. (2012). Me and my 400 friends: The anatomy of
O'Sullivan, P. B., & Carr, C. T. (2017). Masspersonal communication: A model bridging the
Otto, A. K., Laurenceau, J-P., Siegel, S. D., & Belcher, A. J. (2015). Capitalizing on everyday
positive events uniquely predicts daily intimacy and well-being in couples coping with
http://dx.doi.org/10.1037/fam0000042
Fredrickson, B. (2000). Extracting meaning from past affective experiences: The importance of
peaks, ends, and specific emotions. Cognition and Emotion, 14, 577-606.
http://dx.doi.org.augie.idm.oclc.org/10.1080/026999300402808
Reis, H. T., Clark, M. S., & Holmes, J. G. (2004). Perceived partner responsiveness as an
organizing construct in the study of intimacy and closeness. In D. J. Mashek & A. Aron
(Eds.), Handbook of closeness and intimacy (pp. 201-225). Mahwah, NJ: Lawrence
Reis, H. T., Smith, S. M., Carmichael, C. L., Caprariello, P. A., Tsai, F-F., Rodrigues, A., &
Maniaci, M. R. (2010). Are you happy for me? How sharing positive events with others
Reitz, A. K., Motti-Stefanidi, F., & Asendorpf, J. B. (2016). Me, us, and them: Testing
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: University Press.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic
http://dx.doi.org/10.1037/0003-066X.55.1.68
Stavrositu, C., & Sundar, S. S. (2012). Does blogging empower women? Exploring the role of
doi:10.1111/j.1083-6101.2012.01587.x
Sundar, S. S., Jia, H., Waddell, T. F., Huang, Y. (2015). Toward a theory of interactive media
effects (TIME): Four models for explaining how interface features affect user
Tobin, S. J., Vanman, E. J., Verreynne, M., & Saeri, A. K. (2014). Threats to belonging on
doi:10.1080/15534510.2014.893924
Trepte, S., Dienlin, T., & Reinecke, L. (2015). Influence of social support received in online and
offline contexts on satisfaction with social support and satisfaction with life: A
Utz, S. (2015). The function of self-disclosure on social network sites: Not only intimate, but
Vogel, E. A., Rose, J. P., Roberts, L. R., & Eckles, K. (2014). Social comparison, social media,
http://dx.doi.org/10.1037/ppm0000047\
Zell, A. L., & Moeller, L. (2017). Narcissism and “Likes”: Entitlement/Exploitativeness predicts
both desire for and dissatisfaction with responses on Facebook. Personality and
Table 1
Linear mixed models predicting perception of status updates subjects as positive and as important.
Status Update Perceived as Positive Status Update Perceived as Important
Parameter Estimate (SE) t 95% CI Estimate (SE) t 95% CI
Intercept 8.384 (.231) 36.27** [7.928, 8.841] 5.880 (.351) 16.76** [5.188, 6.573]
Level 1 (status update-specific)
Likes .016 (.004) 4.13** [.009, .024] .012 (.006) 2.12* [.001, .024]
Positive Com. -.027 (.016) -1.66 [-.059, .005] .037 (.024) 1.55 [-.010, .083]
Neutral/Negative Com. -.397 (.058) -6.87** [-.510, -.283] -.174 (.083) -2.09* [-.337, -.011]
Participant’s Own Com. .072 (.039) 1.85 [-.005, .149] -.031 (.057) -0.55 [-.143, .081]
Days old -.023 (.016) -1.41 [-.055, .009] -.022 (.024) -0.94 [-.068, .024]
Level 2 (participant-specific)
Age .017 (.008) 1.97 [.000, .033] -.003 (.013) -0.25 [-.028, .022]
Gender .203 (.245) 0.83 [-.280, .686] .479 (.370) 1.29 [-.251, 1.209]
# of Facebook friends .000 (.000) -1.11 [-.001, .000] <.001 (<.001) 0.68 [<.001, .001]
# of status updates -.030 (.015) -2.03* [-.060, -.001] -.046 (.023) -2.02* [-.092, -.001]
Note. All predictors mean-centered, except gender (scored as female = 1, male = 0).
Com. = Comments.
* p < .05, ** p < .01
IMPORTANCE OF LIKES AND COMMENTS 32
Table 2
Linear mixed models predicting recall of status updates and satisfaction with responses received to status updates
Recall of Status Update Satisfaction with Response to Status Update
Parameter Estimate (SE) t 95% CI Estimate (SE) t 95% CI
Intercept .888 (.092) 9.67** [.707, 1.070] 6.755 (.339) 19.92** [6.086, 7.425]
Level 1 (status update-specific)
Likes .007 (.002) 3.82** [.003, .010] .037 (.004) 9.06** [.029, .045]
Positive Com. -.005 (.008) -0.64 [-.020, .010] -.003 (.016) -0.17 [-.034, .029]
Neutral/Negative Com. -.005 (.026) -0.20 [-.057, .046] -.037 (.058) -0.65 [-.150, .076]
Participant’s Own Com. .072 (.025) 2.86** [.023, .121] .101 (.041) 2.44* [.020, .182]
Days old -.078 (.007) -11.13** [-.092, -.064] -.007 (.016) -0.43 [-.039, .025]
Positivity .019 (.014) 1.38 [-.008, .046] .229 (.032) 7.11** [.166, .292]
Importance .009 (.010) 0.95 [-.010, .028] -.047 (.022) -2.08* [-.091, -.003]
Level 2 (participant-specific)
Age .001 (.003) 0.36 [-.005, .008] .021 (.012) 1.75 [-.003, .045]
Gender .086 (.098) 0.88 [-.108, .281] .581 (.350) 1.66 [-.109, 1.270]
# of Facebook friends .000 (.000) -0.58 [.000, .000] <.001 (<.001) 0.05 [-.001, .001]
# of status updates -.029 (.006) -5.01** [-.041, -.017] .024 (.024) 1.03 [-.022, .071]
Note. All predictors mean-centered, except gender (scored as female = 1, male = 0).
Com. = Comments.
* p < .05, ** p < .01
IMPORTANCE OF LIKES AND COMMENTS 33
Table 3
Simultaneous regressions predicting subjective well-being and perceptions of how supportively one’s Facebook
community responds to one’s good news
Passive-Destructive
Subjective Well-Being PRCA-F Composite
subscale of PRCA-F
Zero-order Zero-order Zero-order
β β β
Predictor correlation correlation correlation
Gender (female = 1, male = 0) .01 -.10 .14* .09 -.10 -.04
Age .15** .19** .11 .03 -.23** -.16*
Number of Facebook friends .05 .03 .07 .06 <.01 .02
Number of status updates
.04 -.03 .14 .04 -.19** -.11
posted in last 2 weeks
Number of likes + comments
received on most popular .21** .20** .16* .14* -.21** -.19**
status update in last 2 weeks
F(5, 218) = 4.00** F(5, 218) = 1.83 F(5, 218) = 4.35**
Note. PRCA-F refers to Perceived Responses to Capitalization Attempts scale (Gable et al., 2004) modified for
Facebook. For correlations that involve ratings of specific status updates n = 219-224; for correlations that do not
involve ratings of status updates n = 307.
* p < .05, ** p < .01
IMPORTANCE OF LIKES AND COMMENTS 34
Supplemental Materials
In this supplemental file, we provide additional detailed information about the Perceived
Responses to Capitalization Attempts scale (original: PRCA, by Gable et al., 2004) that we
modified for a Facebook context (PRCA-F).
Factor Analysis. As mentioned in the main manuscript, we submitted the PRCA-F items
to a principal component analysis with varimax rotation, and found the expected four factors.
We provide the SPSS output from that analysis here, for anyone who might be interested in
seeing the factor loadings.
IMPORTANCE OF LIKES AND COMMENTS 35
Intercorrelations among the PRCA-F Subscales. Using the original PRCA in reference
to face-to-face interactions, Pagani, Donato, and Iafrate (2013) found that the Active-
Constructive subscale correlated negatively with the Active-Destructive, Passive-Constructive,
and Passive-Destructive subscales. In our sample using the PRCA-F, the Active-Constructive
subscale correlated negatively with the Passive-Destructive subscale (r = -.14, p = .02) but not
with the Passive-Constructive subscale (r = .26, p < .001) or with the Active-Destructive
subscale (r = .07, p = .23). We did match Pagani et al. (2013) in finding that the other subscales
all correlated positively with each other: Passive-Constructive and Passive-Destructive (r = .25,
p < .001), Passive-Constructive and Active-Destructive (r = .25, p < .001), and Passive-
Destructive and Active-Destructive (r = .35, p < .001).
IMPORTANCE OF LIKES AND COMMENTS 36
The instructions that we provided our participants for the PRCA-F stated, “Please rate
how accurately each of the statements below describe your Facebook experience.” The stem
said, “When I post on Facebook about something good that has happened to me…” and then the
PRCA-F items followed, each rated on a scale from 1 to 7. The Supplementary Table below
shows how each of the individual items correlate with our outcome variables: subjective well-
being, the total number of likes and positive comments received on the participant’s most
popular post, the mean number of likes the participant received on status updates, the mean
number of positive comments the participant received on status updates, and the participant’s
mean satisfaction with the responses to his/her status updates.
(r = .16, p = .02), and mean positive comments (r = .17, p = .01), and mean satisfaction with
responses (r = .20, p = .002).
Supplementary Table: PRCA-F items descriptive information and correlations with subjective well-being, measures of amount of responses received to
status updates, and satisfaction with responses received to status updates
Correlations (r)
Likes + Positive Mean Mean Mean
Subjective Comments on Likes / Positive Satisfaction
Well- Most Popular Status Comments / with
PRCA-F Item M (SD) Being Status Update Update Status Update responses
Passive-Constructive Subscale:
My Facebook friends say little, but I know they are
1 4.39 (1.55) .10 -.17* -.10 -.13* .01
happy for me
My Facebook friends try not to make a big deal out
2 3.44 (1.50) .00 .04 .07 .09 -.11
of it, but are happy for me.
My Facebook friends are usually silently supportive
3 3.88 (1.48) -.05 -.10 -.09 -.11 -.08
of the good things that occur to me.
Active-Constructive Subscale:
I sometimes get the sense that my Facebook friends
4 3.26 (1.57) -.08 -.08 -.04 -.12 -.08
are even more happy and excited than I am.
My Facebook friends usually react to my good
5 5.15 (1.46) .21** .11 .08 .12 .22**
fortune enthusiastically.
My Facebook friends often ask a lot of questions and
6 3.93 (1.62) .16** .20** .17* .16* .12
show genuine concern about the good event.
Passive-Destructive Subscale:
My Facebook friends don’t pay much attention to
7 3.05 (1.37) -.26** -.21** -.10 -.19** -20**
me.
Sometimes I get the impression that my Facebook
8 2.78 (1.62) -.35** -.16* -.07 -.13* -.26**
friends don’t care much.
9 My Facebook friends often seem disinterested. 2.84 (1.48) -.29** -.17* -.09 -.20** -.21**
Active-Destructive Subscale:
10 My Facebook friends often find a problem with it. 1.66 (1.06) -.32** -.07 -.06 -.06 -.21**
My Facebook friends remind me that most good
11 2.28 (1.52) -.13* .12 .17* .22** -.05
things have their bad aspects as well.
My Facebook friends point out the potential down
12 2.03 (1.31) -.17** .11 .16* .18** -.09
sides of the good event.
Notes: For correlations with subjective well-being, n = 307. For correlations with measures of responses, n ranges from 219-224.
* p < .05, ** p < .01