Transcript Facebook Stalking Statusesx

• Select 100 random people
•Our friends in our network, our parent’s network, and friend’s
networks
•Collect data based on people’s statuses
•Compare Males vs. Females
•Compare Young vs. Old
•Young= ages up to 10
•Old= 30+
•Compare statuses
•Relationship
•Song lyrics
•What they’re doing
•Reflections
•Other
•Facebook is a social networking website launched in February 2004
•Facebook was founded by Mark Zuckerberg with his college roommates
and fellow computer science students Eduardo Saverin, Dustin Moskovitz
and Chris Hughes
•The website's membership was initially limited by the founders to Harvard
students, but was expanded to other colleges in the Boston area, the Ivy
League, and Stanford University
•It later expanded further to include (potentially) any university student,
then high school students, and, finally, to anyone aged 13 and over
• The website currently has more than 400 million active users worldwide
•Entertainment Weekly put it on its end-of-the-decade 'best-of' list,
saying, "How on earth did we stalk our exes, remember our coworkers' birthdays, bug our friends, and play a rousing game of
Scrabulous before Facebook?"
•Who wastes the most time on facebook updating statuses
•Males or Females?
•What age demographic updates the most?
•Young or Old?
•What is the most common theme of statuses?
•Relationship, Song Lyrics, What they’re doing, Reflection, Other?
•Time frame
•The people who we check
•If people update a status, then delete it
•Human error
•People may lie about age
•Both female and male graphs
are right skewed
•The center for both is around 1
•Females- min: 0 max:8
•Males- min:0 max:4
•Females have a greater number
of status updates, the most for
females is 8 and most for males is
4. But they have the same center
which means that 8 might be an
outlier.
State
2 independent SRS
2 normal pop
Or
N1 and N2 > 30
Check
Calc random int
N1 and N2 > 30
•Sample count of Gender = Female: 59
•Sample count of Gender = Male: 42
•Alternative hypothesis: The population mean of total where Gender = Female is not
equal to that where Gender = Male
•The test statistic, Student's t, using unpooled variances , is 1.809. There are 98.8309
degrees of freedom.
•If it were true that the population mean of total where Gender = Female were equal
to that of total where Gender = Male (the null hypothesis), and the sampling process
were performed repeatedly, the probability of getting a value for Student's t with an
absolute value this great or greater would be 0.073.
Ho:
m = f
Ha:  m<
f
t
x1  x 2
2
2
s 1 s 2

n1 n 2
Based on our statistics Females updated their statuses more frequently than males did.
•Pvalue:
•P(t>1.809|df=98.8309)=.073
•We fail to reject our Ho because our Pvaule is greater than
α=.05
•We have sufficient evidence that the difference between
the means of Male and Female facebook status updates is
greater than 0.
•Both graphs are right skewed
•The center for old is around 2
and the center for young is
around 3
•Old: min- 0 max- 4
•Young: min-0 max- 8 ½
•Based on these graphs the
majority of people young or
old post no statuses. But it is
clear to see that younger
people update more
frequently than older people.
•Sample count of Age = Old: 39
•Sample count of Age = Young: 62
•Sample mean of total where Age = Old: 0.487179
•Alternative hypothesis: The population mean of total where Age = Old is not equal to
that where Age = Young
•The test statistic, Student's t, using unpooled variances , is -5.846. There are 97.7689
degrees of freedom.
•If it were true that the population mean of total where Age = Old were equal to that
of total where Age = Young (the null hypothesis), and the sampling process were
performed repeatedly, the probability of getting a value for Student's t with an
absolute value this great or greater would be < 0.0001.
Ho:
Ha:
o = y
 o < y
t
x1  x 2
2
2
s 1 s 2

n1 n 2
State
2 independent SRS
2 normal pop
Or
N1 and N2 > 30
Check
Calc random int
N1 and N2 > 30
It is more common for older people to not update their status at all as opposed to younger
people. Also younger people have a great range showing that they update more times that
older people.
•Pvalue:
•P(t>-5.846|97.7689)=.0001
•We reject out Ho because our Pvaule is less than α=.05
•We have sufficient evidence that the difference between the
means of Young and Old peoples facebook status updates is greater
than 0.
4
19
Relationship
35
Song Lyrics
27
What you're doing
Reflections
66
Other
Chi Squared Test
• Assumptions:
• 2 independent variables
• All exp counts greater
than 5
• Check:
• Random int. in calc
• checked
Chi-Square
(obs  exp)
x 
exp
2
2
(4  15.1) 2 (35  75.5) 2

15.1
75.5
 106.8
p( x  106.8 df  4)  3.47 10
2
4
Conclusions
• We fail to refect HO because p-value is less
then .05
•Young people, especially teenagers update their status more
frequently than older people
•The most common status update is “What you’re doing”
•Females are more likely to update their statuses than males
•Just because you have a Facebook does not mean you have to
update your status every second
•Some people have never touched their status
•It was annoying going through a 100 people’s status
•No one really cares what you are doing, so all the status updates
are not necessary
•Teenagers will most likely grow out of the updating phase
because older people don’t update as much