Professional Documents
Culture Documents
COMM88
Fall 2018
Since human beings are social groups, people are inevitable to enter an unfamiliar
environment in everyday life. In this case, the initiation of conversation and communication
individual choose to sit close to and talk to a person instead of another person? What factors
are playing roles when people choose the person to sit close to? Is physical resemblance
affecting the seating choice in a room full of strangers? Series of experiments are performed
to investigate whether physical similarities have influence on seat choice when people enter
an unfamiliar environment.
Researchers have been interested in stranger interaction for a few years. A few
researches have been conducted to answer the questions above. However, most researchers
put emphasis on variables such as proximity and similarities in attitude and values rather than
physical similarities. One of the researches related is a study Mackinnon, Jordan, and Wilson
(2011) conducted to investigate how two physical similarities, glass wearing and sex, exert
effects on people’s choice of seat. In the study, seating arrangement was observed at a
computer lab with 31 seats in the library at a Canadian university. For each occupied seat in
the lab, the researcher recorded (a) the sex of the person sitting and (b) whether the person
was wearing glasses. Observations were recorded 1-3 times daily on weekdays, most
frequently during the midday, which was when the library computer lab was most populated.
The result shows that glasses wearers will tend to sit beside other glasses wearers more
frequently than expected by chance and women will tend to sit beside other women (and men
by men) more frequently than expected by chance. To verify the effect of sex, or say, gender
H1: Individuals are more frequent to sit close to people with the same gender.
3
Besides gender, there are multiple aspects of similarity. Luo (2009) performed a
research to investigate the role of couple similarity in a sample of 117 couples who recently
started dating. During their visit to the lab, participants completed a packet that includes a
interests, and various personality characteristics. The result is that for interests, couples
showed a significant correlation on sport activity and habitual activity. In this case, it is
reasonable to assert that couples may show a significant correlation on body shape because
body shape, generally speaking, reflects sport activity and habitual activity. It is widely
acknowledged that approaching and conversation are first two steps of a relationship, which
leads to
H2: Individuals are more frequent to sit close to people with similar body shape.
Method
In order to examine the effect of gender and body shape when people choose seats in
contain 8-12 persons in the library of University of California, Santa Barbara is booked as
test field. The same chairs are put as a circle in the room. All chairs are divided into groups of
two and adjacent two chairs are in a group. All the distances between two adjacent groups of
chairs are similar. Participants do not know any volunteer who sits in the room. Participants
are asked to choose a chair to sit in the room and to complete a short questionnaire. In this
way, the real purpose of the experiment is concealed and personal information of participants
Sample
In the experiment, convenience sample is employed to collect data. Participants are randomly
selected from people who study in the library. They are asked whether they are interested in a
communication experiment or not to make sure they are voluntary to participate in the
4
them are female, while 9 of them are male. To test hypothesis 2, two experiments are
conducted. The first experiment has 19 participants, among whom are 12 males and 7
Procedure
Before the experiments were performed, a pre-test was conducted. Among the twelve chairs
in the room, the six chairs which volunteers would sit later were not permitted to sit on.
Participants were free to choose any of the seat except for these six chairs. The seat each
participant chose was recorded. The result reveals that no chair is particularly popular. Every
In the experiment which investigates the influence of gender, all chairs in the room were
divided into six groups. Six volunteers, three males and three females, sit on chairs of six
different groups before each participant entered the room. Each seat had a vacant chair close
to it to make sure that each participant had no choice but to choose to sit next to one of the
volunteers. Each volunteer held several copies of questionnaire which contains basic
questions such as gender, height, and weight. All volunteers were similar in other physical
appearances except for gender. Participants were asked to choose a seat, gain a copy of
questionnaire from the volunteer next to them, and complete the questionnaire on the seat. In
first experiment which examines the effect of body shape, six volunteers are employed, three
males of different body shapes and three females of different body shapes. In the second
experiment, all volunteers are four males of different body shapes. All of other procedures are
Measures
The dependent variable of the experiment is whether people sit close to people with same
gender and similar body shape. Frequency and percentage are appropriate tools to measure
5
the dependent variable. After the participants finish the questionnaire and leave, who they sit
close to is recorded on the questionnaire. To quantify body shape, BMI (body mass index) is
utilized. According to National Heart, Lung, and Blood Institute, body mass index (BMI) is a
measure of body fat based on height and weight that applies to adult men and women.
To define what means similarity in body shape, the difference of BMI is used. BMI normally
has four categories – underweight, normal weight, overweight, and obesity. Although there
are some subtle differences between divisions of categories on internet, the categories of
normal weight and overweight tend to have range differences about 6. As a result, a
difference below 6 can be considered as an apt measure showing that a participant and a
Results
comparisons between different variables. H1 predicts that people are more likely to sit close
to people with the same gender. Analyses of frequency of how many individuals sit beside
people with same gender support this hypothesis. Table 1 demonstrates that 5 female
participants of 14 all sit close to female volunteers while 9 male participants all sit close to
male volunteers. The frequency is 14 and percentage is 100%, which is able to support the
H1. H2 predicts that people are more likely to sit close to people with similar body shape.
The first experiment discusses about BMI of same gender. Table 2 shows that 9 male
participants of all 12 males sit next to male volunteers whose BMI has a difference below 6
from them, which means 75% (9/12=0.75) of male participants choose seats next to people
with similar body shapes. It is also noticeable that among 7 female participants, 6 choose
6
female volunteers who have similar body shapes, which leads to a percentage of 86%
(6/7=0.86). Therefore, in the case of same gender, H2 is supported. Whether body shape
According to table 3, 83% (5/6=0.83)female participants sit close to male volunteers who
have similar body shape with them. Hence in the case of different genders, H2 is supported as
well. Then, it is not difficult to draw the conclusion that H2 is supported no matter gender.
Discussion
The results illustrate that college students are more probable to sit close to people with
the same gender and similar body shape when they enter an unacquainted environment. All
these are consistent with the study of Mackinnon, Jordan, and Wilson and Luo. Due to the
characteristic of experiment, the external validity is not satisfactory enough, which means
that these data are of limited generalizability. This may be true if compared with surveys.
However, if so, it is equally true for most other studies. Data in this experiment are collected
in an actual field setting – a library, which is one of the scenes of everyday life. Compared to
the results of other studies that are performed in entirely experimental environments, these
results have higher external validity because the actual field setting enables naturally, instead
The largest limitation of these data, however, lies in internal validity. Because of the
restriction of time and location, convenience sample has to be employed. All participants are
college students who study or have a rest in the library. When selection of subjects is not
really a random sample, selection bias exists. In such case, one of the most significant parts
of true experiment is lack – random assignment, which is also limitation of majority of quasi-
experiments. This is the reason why this experiment cannot contribute to causality. Instead,
the only conclusion researchers can draw is that gender and body shape of volunteers have
strong relationship with seat choice of participants. To achieve a causal conclusion, richer
7
data of different groups of people in the society are essential. The other important element of
internal validity is to control the environment well in order to eliminate other explanations. In
the first experiment to examine BMI effect, as a result of time restriction, three male
volunteers of different body shapes and three female volunteers of different body shapes have
to be in the room at the same time, which means, as a matter of fact, two independent
variables are included in this round. This can cause an extreme descent in internal validity.
Nonetheless, if the result of former experiment that 100% participants sit next to volunteers
of the same gender is taken into consideration, the serious consequence of result error can be
evited since gender diversity will not affect the choice of BMI. Hence, the result can still be
persuasive and compelling. However, for the sake of safety, this experiment is supposed to be
divided into two sub-experiments, one with 3 male volunteers of different body shapes and
almost all pre-test/post-test experiments have also unavoidably exists in this experiment. The
problem is that differences in data of dependent variables could be the result of interaction of
manipulation with pre-test. To be more specific, when participants enter the room for the
second time, they are likely to sit on the same chair they have sit before regardless of who sits
environment and be more comfortable in where they are familiar. Due to the lack of Solomon
4-group design, the comparison of pre-test/post-test group and post-test only group cannot be
made so it is difficult to determine whether pre-test has an effect on seat choice. To solve this
The experiment also contains a questionnaire to collect the personal information of all
height, and weight, all of which are open-ended questions. All the questions are succinct and
8
pellucid to make sure that there are no other interpretations of questions. Under this
doubt the authenticity of information because factually inaccurate response may be gathered.
It is likely that participants do not know their accurate height and weight and report an
approximate number. Or the height and weight they know have changed and are different
from the current data. The result of BMI calculation is likely to be impacted. As a result, error
Clearly, other physical traits such as hair color and race need to be examined to more
adequately. More data of samples are supposed to be gathered and the experiment should be
replicated to achieve higher reliability. In addition to the methodical problems stated above,
there are some other social and practical issues. In the contemporary society, gender cannot
be briefly categorized as male and female. A few individuals have gender identity and
expression that does not align with male and female. Those people are ignored automatically
in this experiment. Another problem is that sometimes subtle difference in BMI cannot show
evident difference in body shape. From table 2, one male volunteer has BMI of 18.4 and one
male volunteer has BMI of 19.1. These two numbers are close and the difference is hard to be
told if participants just use their eyes to make a decision. It is difficult for researchers to find
volunteers whose body shapes are broadly different. But because the experiment investigates
similarity, the impact of two volunteers with similar body shape can be omitted. From table 2,
(26.2-17.2=9). Because the range of BMI is large enough, at least body shapes of two
which means it can predict the outcome of a similar situation based on the study findings.
When a college student enters a room full of strangers, where he or she tends to sit and who
9
he or she tends to initiate a conversation with can be predicted. This strategy can be applied
to considerable fields such as advertising and marketing. Future studies can further
investigate the topic of seating choice. Will people with physical resemblances continue to
talk and talk more? Will they develop their relationship more rapidly than those with rare
physical resemblances? Are they conscious or unconscious to make such choice? All those
References
https://www.nhlbi.nih.gov/health/educational/lose_wt/BMI/bmicalc.htm
Luo, S. (2009). Partner selection and relationship satisfaction in early dating couples: The
doi:10.1016/j.paid.2009.02.012
Mackinnon, S. P., Jordan, C. H., & Wilson, A. E. (2011). Birds of a Feather Sit Together:
Table 1
Table 2
Table 3
Appendix
Abstracts
Across four studies, people sat (or reported they would sit) closer to physically similar
others. Study 1 revealed significant aggregation in seating patterns on two easily observed
characteristics: glasses wearing and sex. Study 2 replicated this finding with a wider variety
of physical traits: race, sex, glasses wearing, hair length, and hair color. The overall tendency
for people to sit beside physically similar others remained significant when controlling for
sex and race, suggesting people aggregate on physical dimensions other than broad social
categories. Study 3 conceptually replicated these results in a laboratory setting. The more
physically similar participants were to a confederate, the closer they sat before an anticipated
interaction when controlling for sex, race, and attractiveness similarity. In Study 4, overall
physical similarity and glasses wearing similarity predicted self-reported seating distance.
These effects were mediated by perceived attitudinal similarity. Liking and inferred
Although much research has been conducted to test the role of couple similarity in
partner selection and predicting relationship satisfaction, previous research was limited to
14
sample of 117 couples who recently started dating (dating length 6 six months). The early
dating couples showed strong to moderate similarity on demographics (age and ethnicity),
personal interests, values, and political attitudes but weaker similarity on a range of
personality domains. With regard to satisfaction prediction, the absolute difference tended to
be a better predictor than the profile similarity. However, couple similarity overall was not a
strong predictor.