Descriptive Statistics and Regression Analysis

Introduction
The business taken into account for the analysis is Google, Inc. The primary data being collected is on Google and on different mobile applications to evaluate the validity of whether the latter are becoming more significant and popular compared to Google. In this case, the independent variable is percentage of internet usage on Google for every period. Basically, the metric taken into consideration is the percentage of time spent on Google in a day against the percentage used on various other Mobile Applications. This is purposed to facilitate the establishment of trends between Google and Mobile Applications and attain a vivid picture of their individual popularities.
Descriptive Statistics
Descriptive statistics takes into account the processes employed to provide a summarization and delineation of the significant characteristics of a set of measurements. Descriptive statistics provides basic sum-ups regarding the sample and the measures and in tandem with basic graphical examinations, it constitutes the foundation of almost every quantitative analysis of data (Mendenhall et al., 2012). There are different types of descriptive statistics that might be ideal for summarizing the data, if a sample of the Google data were to be collected. First, there is there is the distribution, which is a summary of the regularity or rate of recurrence of individual values or ranges of values for a variable. The most basic and simplest distribution would describe every value of a variable and the number of individuals who had every value. For instance, in the case of Google, a typical way to describe the distribution of popularity is by age of individuals, listing the frequency of use of Google or mobile applications. In the same manner, the data could be described by gender through the listing of the number or proportion of males and females. This data can be described using a frequency distribution, which can be portrayed as a table or as a graph (Trochim, 2006).
The second type of descriptive statistics that would be best for summarizing data is the central tendency. In delineation, the central tendency of a distribution is an approximation of the center of a distribution of values. These estimates include mean, median, and mode. The mean provides the average value of the data and is obtained through the summation of all the values and dividing them by the number of values. In this case, the mean will provide information regarding the average number of internet users that use Google or use mobile applications in between the period of observation and analysis. Secondly, the median is the value obtained at the precise middle of the set of values. Third, the mode is the most regularly occurring value in the set of values. The mode shows frequency. Therefore, in this case, it will be possible to determine important characteristics such as the recurring age of the users that prefer either Google or mobile applications or also determine the rate of usage of either the two (Trochim, 2006).
Inferential Statistics
Inferential statistics comprises of processes employed to make inferences regarding population characteristics from data and information contained in a sample obtained from a population. The key purpose of inferential statistics is to draw conclusions, make forecasts, as well as make decisions regarding the characteristics of a population from information that is contained within a sample (Mendenhall et al., 2012). Through inferential statistics, it becomes possible to reach conclusions that go beyond solely the instantaneous data. Therefore, on one hand, inferential statistics is employed to make inferences from the data collected to more overall circumstances whereas descriptive statistics is employed to basically delineate what is going on with the data (Trochim, 2006). One type of inferential statistics that might be ideal for analyzing the data is the t-test. Basically, the t-test examines whether the means of two groups are statistically dissimilar from one another. In this case, the approach will be deemed suitable in comparing the means of the two groups being considered. Another kind of inferential statistics that is ideal in this case is the analysis of variance (ANOVA) approach, which facilitates the analysis of the differences amongst the means of two different groups (Trochim, 2006). Therefore, through these inferential statistics, it would be possible to infer and make conclusions regarding the population of internet users at large from the sample that will be employed in the analysis.
Probability or Trend Analysis
Trend analysis plays a fundamental role in impacting the process of comparing dissimilar data and information. Imperatively, the practice of researching data and information together with the presentation of outcomes is largely dependent on the outcomes together with the constancy of the trends in the collected information. In this regard, trend and probability analysis will make it possible to comprehend the inclination of internet users towards the use of Google and mobile applications. In the same manner, trend and probability analysis will help in predicting whether the prevailing inclination will give rise to the increase or decline in popularity of Google amongst users. Therefore, trend analysis can be employed to enhance Google as an organization through determining the different aspects where Google beats mobile applications, where it is underperforming and also provide information to facilitate decision making. Moreover, probability will play the role of delineating the prospective results of a certain event or course of action in addition to the statistical likelihood of every event. Therefore, in this case, probability can be employed to mirror the likelihood that Google will cease being popular or whether Google will continue to surpass the popularity of mobile applications (Brandimarte, 2012).
Linear Regression for Trend Analysis
Linear regression is a procedure employed to ascertain the magnitude to which there is a linear relationship or correlation between a dependent variable and at least one or more than one independent variables. Basically, in this approach, an independent variable is utilized in the prediction of the value of a dependent variable. In this case, linear regression for trend analysis would play a significant role in helping address the business problem. Notably, the notion of regression would be to examine whether an independent variable such as internet usage can predict the popularity of Google as the dependent variable. The linear regression analysis would be employed to explain the correlation between internet usage and popularity of mobile applications and Google. This would include determining whether the increase in the usage of mobile applications amongst internet users will give rise to a decline in the popularity of Google (Mendenhall et al., 2012).
Time Series
Time series analysis takes into account approaches for examining time series data so as to obtain sensible statistics together with other characteristics of the data. In simple, through time series forecasting, it is possible to predict values in the future on the basis of preceding values observed. In this case, time series will play a key role in ascertaining the nature of the popularity of Google as compared to that of mobile applications characterized by the sequence of observations and also in forecasting the popularity of both Google and mobile applications. Through time series, it becomes possible to extrapolate the ascertained pattern of the data for Google in order to make a prediction of future events (Mendenhall et al., 2012).
References
Brandimarte, P. (2012). Quantitative methods: An introduction for business management. Hoboken: John Wiley & Sons.
Mendenhall, W., Beaver, R. J., & Beaver, B. M. (2012). Introduction to probability and statistics. New York: Cengage Learning.
Trochim, W. M. (2006). Descriptive Statistics. Web Center For Social Research Methods.