Abstract
There is always a need for the analysis of various data and information found about an activity to determine how several things are influenced by this data. This data can be for the genearal population or different people where the nature of activy data will be varying. The data and other sampled output will alws assist the researcher to study some of the specified behaviour in the subjects. The first step in the analsysis of any given data will be to understand what the data is alll about. Then the goals of the study or analysis of the data is done and this will ensure that the researcher does not stray fromhis objectives. The analysis of the dat is done and this can be in terms of various data analysis techniques like regression, charts.
Introduction
The improtant of text messaging cannot be over-emphasized in the current communiacation realm. Text messaging has taken over all other traditional source of communication as the numebr one source of information sharing. This medium of communication is convenient and due to the fact that a large percentage of people. The use of text massaging varies with different factors like the gender and age. This use has increased significantly over the past year and it has led to various effects on the general population sociobility.
It will be of great improtance to find the figures and numbers on the usage of text messaging as a means of communication. This data found will be of great importance in the analysis of the trend in messaging, will assist us make an informed decsion on the different aspects of text messaging. This data will give us a wider view on the benefits and effects of text messaging. The data collected can be stored for future referencing and analysis and thus it is of great importance to find these data.
Goals
The goals of the study and analysis of text messaging are as follows:
-The first goal is to compare the average of amount of time spend doing homework and the time spend hanging out with friends.
- The second goal is to do a detailed comparison between the males and the females usage of text messaging in terms of volumes of text messages received and sent, time taken to hang out with friends and doing homework.
-The third goal is to find the correlation between any of the given variables of choice and in our case is to find the correlation between the number of texts received and the number of hours spend hanging out with friends.
-The fourth goal is to find the mean population estimate of the number of texts received, the number of texts received and the time taken to hang out with friends for any sample in the data and the accuracy in doing so.
Variables in data
The different variablea in the data are numebr of hours taken to hang out with friends and homework hours, the number of texts send or received.
Analysis of the data
Using descriptive statistic, there are several realisations on the data on the amount of time used to do do homework and hang out with friends. This can be as below:
The total number of hours spend hanging out with friends for the two groups is 1822 hours with the mean of the time being 18.22 hours. On the other hand, tht total number of hours of spend doing homework is 928.5 hours with a mean of 9.37 hours.
Using pivot table in the analysis, the following data was obtained which could be analysed to get the differences between male and female in the messaging activities.
The female and male messages send and received and the time thay all take to hang around with friends and to do homwork is hughlighted in the pivot table above.
For the regression analysis of the given data, the procedure of the statement of the problem, the slection of the variables, data collection, specification of fitting method and model and model validation and criticism of the given data will be performed . this will give us the given functional relationships between the given entities.The relationship in our case is whether the text send, received and the time for homeowrk or to hang out with friends are related in any way. The corellation between these data is done by the use of scatter plots. We are determining the correlation between the number of hours spent hanging out with friends in regards to the text messages send
The table for the computation of correlation coefficient is in the appendix: It gives a coreelation coefficient of 0.333.
For the test of the hypothesis of whether the answer is right and ther is a coreelation between the text messages sent and the number of hours spent, the following is tested for a two tailed distribution
t=r1-rn-2 where r is 0.33 in our case and n is the number of variables which are 100
This will give 0.331-0.33100-2 = 391
For the estimation of population, we take the confidence level of the population to be given by 0.95. The critical value for this confidence intervalis given by 1.96 in the statistical tables. The estimation of the the mean number of texts send and received, and the number of hours spend hanging out with friends and doing homework at any given time can be done as below:
x±1.96(σn) The standard deviation is done for the number of text messages received sample from the appendix as the differences of the means then squaring the means and finding the average of the mean which gives the variance. The standard deviation of the data is found by applying the following formula on the excel with the data and specifying the data range:
Standard deviation = STDEV.S(Column1:Column n)
The various standard deviations calculated from the excel gives the folowing results.
The standard deviation chart can be shown below.
Estimation calculation for message sent is 1198.82±1.96(8808.25100)
Which is equal to 1198.82±1726.417 =(-527.597, 2919.237)
For the time spent hanging out with friends, the standard deviation is 20.212
Thus the estimation of time spend on hanging out with friends is given by with a confidence interval of 95%, where the mean of the time is 18.22 with the number of samples being 100. 18.22±1.96(20.212100) =18.22±3.9616=(14.258,22.1816)
For time spend doing homework with a mean of 9.379 and standard deviation of 7.47857, estimation is 9.379±1.96(7.4786100) =9.379±1.466=(7.913,10.845)
Conclusion
The average of the amount of time spend doing homework is 9.37 hours while that for the time spend hanging out with friends is 18.22 for both the male and females. It can be concluded that the amount of time spend on doing homework for the two groups is less than the amount of time spend hanging out with friends. Thus the first goal of comparing the amount of time spend doing homework and that spend hanging out with friends has been achieved.
The female do more texting than the males. They also spend more time hanging out with friends and doing homewrok than their male counterparts. This is shown in the pivot table in fig 1 where for example female sent and receive 69183 and 99716 text messages respectively while the male only sent 14660 and 29166 messages. From the fig 2 of histogram comparison between the text messages send and received by the two genders. Thus the second goal of doing a detailed comparison of the activities done by the two genders based on the data has been achieved.
As from the scatter plots of fig 4 and 5, it can be seen that there is a low correlation between the number of text messages received and the number of hours spend hanging out with friends. The plots are too random and no relationship between the given entities can be drawn from it. Thus the messaging behaviour of each individual is independent on the amount of time spend hanging out with the given friends. Thus the third goal of finding the correlation between the numbers of texts received and time hanging out with friends is achieved.
The estimation of the population sample mean at any given time for the number of text messages received was computed with a confidence interval of 95% and it was found to lie within the interval of (-527.597, 2919.237. This shows that the estimation of the mean of number of texts sent cannot be done due to the wide variation of the values. It becomes hard to determine the exact interval where the estimations can lie within the population. This is also represented in figure 10 for the standard deviation. For the estimation of the number of hours spend on hanging our with friends, it can be satisfactory estimated that the mean of any given sample will lie within the interval (14.258,22.1816) with a 95% confidence level. Thus it is easy to estimate the number of hours spent in hanging around friends but very difficult to do so for the number of text messages received. The estimation of mean of number of hours spend doing homework with a confidence interval of 95% gives (7.913,10.845) thus this can be easily estimated from any given sample with a high level of accuracy. This can be shown also on figure 9 showing a small deviation. Thus the goal of finding the ranges of the mean estimate of any selected sample is achieved.
Appendix
Correlation coefficient=(-(-71665640293.49*7.67E+09=0.333
References
Chatterjee, S. a. (2006). Regression analysis by example (4th ed.). New Jersey: Wiley.
Moore, D. N. (2013). The basic practice os statistics (6th ed.). New York: W.H. Freeman and Company.
s
