Tuesday, April 12, 2011

Stat final

Joy Prior
STAT 121 Sec 045
State: Is there a significant difference between the mean value of the amount of pictures on Facebook that Joy Prior’s female and male Facebook friends publish on their profile pictures?
Plan: This is an observational study, and the observations were collected from Facebook an online social network. The test procedure will be a two sample t-test that will compare the parameter of the mean values of the amount of pictures the male members on Joy Prior’s Facebook friend list, and amount of pictures the female members on Joy Prior’s Facebook friend list post on their profile page. The null hypothesis is that there will not be a significant difference between the means. The alternative hypothesis is that there will be a significant difference between the means. The hypothesis aim to discover if there is a significant difference between the mean value of the amount of pictures on Facebook that Joy Prior’s female and male Facebook friends publish on their profile pictures? The set level of significance for this study is alpha equals .05.
One of the limitations of the study could be in the collection of the Simple Random Sample. The sample was taken from a list of all of Joy Prior’s Facebook friends assigned a number 001 to 635 in alphabetical order. To create a random sample the data was collected from 104 Facebook profile pages chosen by a list of random numbers taken from Table C (the random number table provided on blackboard). Another limitation of the study was that the sample was only taken from Joy Prior’s Facebook friend list. This study could not be used as a generalization for all members of Facebook, but could simply be used to describe the specific parameter. One observation of the population of Joy Prior’s Facebook friends is that the majority appeared to be in their early to mid-twenties and did not accurately represent the entire social network. It is important to recognize that there was a difference in the sample size of males and females. The number of females sampled was 71, but the number of males sampled was 33. This disturbance in the data collection was possibly due to the observation that Joy Prior has more female Facebook friends than male Facebook friends.
Solve:
Male

Female

The conditions of the two sample t test included a simple random sample and normality in the sample. As mentioned before one of the limitations of the study was the numbers assigned to the population in alphabetical order and not randomly. The data was collected by a sample using random numbers. There was no normality in the distribution. Both populations were right skewed. According to the inner quartile range and the 1.5*IQR rule the male sample did not have any apparent outliers but the female sample has 3 apparent outliers. Under this situation the conditions are not meet.
The calculated sample statistics:
Male
N=33
Mean= 268.0606
Std. Dev.= 226.70271
Variance= 51394.12
Std. Err= 39.46388
5 Number Summary: Min= 6; Max= 1045; Q1= 119; Mean= 268.0606; Q3= 395
Female
N= 72
Mean= 395.84506
Std. Dev.= 372.7559
Variance= 138946.97
Std. Err= 44.237988
5 Number Summary: Min= 1; Max= 1544; Q1= 80; Mean= 395.84506; Q3= 540
Test Statistics:
T-Stat= -1.8165619
Computer calculated p-value= 0.0722
Computer calculated degrees of freedom= 102
Computer calculated std. err.=70.344124
Computer calculated sample mean= -127.78446
Degrees of Freedom= 32
P-value from Table C= .05< p-value < .10
Conclude: Because the .05< p-value < .10 is greater than the set alpha of .05 we can fail-to-reject Ho and concluded that there is not a significant difference between the mean values of the pictures males post on their profile page and the mean value of pictures females post on their profile page taken from the members of Joy Prior’s Facebook friend list.
Implications: The implications of this data can be used for the organization Facebook to know which features, specifically the tagging pictures, are being used by which group of consumers. While collecting the data it appeared that there are other variables that would influence the amount of pictures someone has on their profile page more directly than male or female. Some example hypotheses are the person’s age, the amount of friends, or the hours someone spends of Facebook. The actions resulting from the study is that Joy Prior will no longer have a stereotype of the amount of pictures females and males publish on their profile pictures.

No comments:

Post a Comment