Evaluating data from a variety of sources using different methods of analysis.- 8 -2.1 Descriptive, exploratory, and confirmatory.... 8 2.2 The strengths and weaknesses of every analysis
Introduction
Statistics is an indispensable tool in research and practical work Today, the use of statistics has expanded far beyond the first starting point of serving the government. Organizations and individuals use statistics to understand data and make decisions. The use of any statistical method is correct only when the study population satisfies the necessary mathematical assumptions of the method Companies today have been applying statistical methods in business to make statistics of resources and information systems My company is planning to improve decision-making and information management through statistical methods As a research analyst I am required to demonstrate my understanding by evaluating and analyzing business data through transformative statistical methods.
This report aims report is to analyze and evaluate business and economic data/information using several statistical methods.
In this report, I will first analyze the nature and processes of business and economic data/information from different sources Some definitions of data, information, and knowledge will be given In addition, the way data can be turned into information and information into knowledge will be analyzed through examples Next, I will evaluate data from various sources using three analytical methods (Descriptive,exploratory, and confirmatory) Then, I will outline the strengths and weaknesses of these three methods A table comparing the differences between the three analytical methods will also be provided Next, I will analyze and evaluate qualitative and quantitative raw business data from a series of examples using appropriate statistical methods.
Statistics methodology in business and economic
Data, information, and knowledge
Data are the facts and figures collected, analyzed, and summarized for presentation and interpretation All the data collected in a particular study are referred to as the data set for the study.( Anderson, D.R ,et al.,2016)
Data includes expressions used to reflect the reality of the research subject Most of these manifestations are measured or observed values of the research variables. These expressions include numbers, words, or images ( Hoang, T and Chu Nguyen Mong, N., 2008)
All these expressions, if put in a non-specific context, will not make sense So, for data to make sense, it must be placed in a specific context for that data to make sense.
Example : An example of a data set with numerical attributes
Length Height Width Weight Quality
Information is the result of processing, arranging, and organizing data so that the reader has more understanding and knowledge In other words, it is the content of the collected data.( Hoang, T and Chu Nguyen Mong, N., 2008)
Information is a concept that has many different meanings, from everyday life to the technical environment Generally speaking, the concept of information is closely related to the constraints of information, communication, control, data, form, instruction, understanding, meaning, thought stimulation, forms, perception, and presentation ( Hoang, T and Chu Nguyen Mong, N., 2008)
Knowledge is what is known Like other related concepts such as truth, belief, and wisdom There is no single definition that is agreed upon by all scholars, but there are many different and still debated theories about the nature of knowledge.Knowledge accumulation is a complex cognitive process: perception, learning,communication, association, and reasoning The term knowledge is also used to imply certain knowledge about something, which can be used to accomplish a certain goal.( Hoang, T and Chu Nguyen Mong, N., 2008)
How data can be turned into information and information into knowledge
The numbers 1.4% , -8.2% , -19.1% , -22.7% , -27.3% are considered data and in this case they are all significant.
The data we collect is the decline/growth in smartphone sales in Quarter 12020 of 5 brands: Xiaomi, Apple, OPPO, Huawei, Samsung After the process of sifting and processing the collected data, we know that out of the 5 phone brands, only Xiaomi emerged as the only brand that can record sales growth On the other hand, Samsung, Huawei, Apple, and OPPO, the other four brands in the industry's top 5 suppliers for Q1 2020 - recorded a decline in sales, with a double-digit drop for most of these brands.
The knowledge we can draw is that Samsung, Huawei, Apple, and OPPO phone companies need to evaluate their business situation and develop an effective business strategy for sales to grow again.
1.2.2 Example 2: Statistical table of the number of candidates achieving 10 points in the exams in 2018 and 2017
Table1:Statistical tableofthenumberof candidatesachieving10pointsin the examsin2018and2017 Sources:LaborNewspaper,2018 The numbers
The data we collect is the number of candidates who scored 10 in subjects in 2018 and 2017 After screening and processing data, the information we have is that the
Number of 10 points in the national high school exam in 2018
Number of 10 points in the national high school exam in 2017
2018 exam may be more complicated than that of 2017 so the score of 10 in 2018 is significantly reduced compared to the 2017 national high school exam.
The knowledge we can draw is that teachers of schools across the country need to re-identify the source of knowledge that the Ministry of Education uses to make exam questions Then create a better lesson plan to help students have better test results in the next test.
Figure4:Example2 :DIKmodel 1.2.3Example3:NumberofinternationalvisitorstoVietnamby meansof arrivalin2018and2019
Table2:NumberofinternationalvisitorstoVietnambymeansof arrivalin2018and2019Sources:General StatisticsOffice
In Table 2 , the numbers are 12.485,000 ; 215,300 ; 2.797,500 ; 14.377,509 ; 264,115 ; 3.366,967 are considered data
The data we collect is “Number of international visitors to Vietnam by means of transport in 2018 and 2019”
After the process of screening and processing data, the information we have is that the number of international visitors to Vietnam by means of transport in 2019 increased compared to the same period in 2018.
The knowledge we can draw is that, although at the beginning of 2019 there was a Covid-19 epidemic in Vietnam However, the number of international visitors still comes to Vietnam a lot This shows that Vietnam's tourism industry is developing very well.
The fourth strength is provides high quality data: allows the research to be conducted in the respondent’s natural environments , ensuring that honest , high quality data is collected.(QuestionPro,2021)
The fifth strength is quick to perform and cheap As the sample size is generally large in descriptive research, the data collection is quick to conduct and is inexpensive.(QuestionPro,2021)
The last strength is forms the basic for decision making The data collected in descriptive research is robust and represents a large population , which makes it easier to analyze and use for decision making (QuestionPro,2021)
The major weakness of Descriptive analysis is limited in that they only allow you to make summations about the people or objects you have actually measured You cannot use the data you have collected to generalize to other people or objects (i.e., using data from a sample to infer the properties/parameters ) (Davis,B,2021) There are no generalizations about the data, and the results aren't 100% correct.
The major strength of the exploratory mode is that it permits the un- structured examination of data and transforms the act of data analysis from a scientific search for mathematical results into an artistic search for empirical patterns It permits any manipulation of the data, but produces results that are simple to understand.Exploratory techniques may involve sophis- ticated computer routines, but never involve elaborate mathematical results The exploratory mode brings the analyst close to the data since the results arise from the mind Data structure is both exposed and imposed by the analyst The inferences made by the analyst are informal and open to criticism Each analyst must reach his or her own conclusions and these conclusions are open to question by any other analyst.(Mayer, L.S., 1980)
The major weaknesses of the exploratory mode are that it does not yield statements that are inferential in the statistical sense or replicable in the scientific sense. Although the analyst uncovers patterns in the data which are suggestive of hypotheses that could be tested in future studies, the analyst is not free to test these hypotheses on the data used to generate them Con- sequently, the exploratory mode is best viewed as the first step in a sequential analytic process It suggests hypotheses which are tested in subsequent analyses These hypotheses are often improved by rough confirmatory analy- sis and finally tested by the confirmatory mode Thus, unlike the confirmatory mode, the exploratory mode cannot stand alone as a method of inquiry(Mayer, L.S., 1980)
A second weakness of the exploratory mode is that it usually benefits from more extensive data than that required by the confirmatory mode Pro- vided certain assumptions are made, the confirmatory mode can study aggregate units in order to model the behavior of individual sampling units (consumers) The exploratory mode has no provision for substituting aggre- gate for disaggregate data Although the data on aggregates can be explored, the exploratory analyst prefers to examine data on individual units if the behavior of the individuals is the topic of analysis This preference often leads to the gathering and management of a large amount of data.(Mayer, L.S., 1980)
The major strength of the confirmatory mode is that the results obtained are replicable and optimal in well-defined areas Given the same theory, models, and hypotheses, two investigators should obtain the same parameter estimates and the same degree of evidence in favor of each research hy- pothesis Ideally, their results would differ only to the extent that they give different substantive interpretations to the statistical inferences made Furthermore, should they offer different statistical results, then the optimality theorems of classical statistics often provide justification for preferring one set of results over the other In areas such as multivariate analysis
- 15 - of variance, where several testing procedures compete but few optimality results obtain,the optimality and replicability of the inferences made are reduced and, consequently, the power of the confirmatory mode is impaired In such areas statisticians are actively searching for appropriate optimality theorems.(Mayer, L.S., 1980)
The second strength of the confirmatory mode is that it yields indicators of the uncertainty associated with its inferences As examples, the estimated standard error is an indicator of the uncertainty of a parameter estimate and the critical level and power are indicators of the uncertainty associated with a hypothesis test. Without the confirmatory mode the data analyst might be able to make inferences about the population being sampled, but these infer- ences would be informal in the sense that no indicator of the uncertainty of the inferences could be given Such inferences would be significantly weaker, in the scientific sense, then the inferences of the confirmatory mode.(Mayer, L.S., 1980)
The major weaknesses of the confirmatory mode are that it requires specification of the theory, models, and hypotheses prior to the examination of the data. Confrontation with data allows the analyst to modify the model and propose alternative hypotheses, but classical statistical methods are designed to provide optimal estimators and tests and are not specifically designed to be aids in the reformulation of the theory, models, or hypotheses Once the model and hypotheses have been changed, confirmatory methods are not optimal for estimating the model or testing the hypotheses In fact, there are almost no theorems in classical statistics which guarantee the optimality of a confirmatory method if the model or hypotheses are re- formulated after an initial confrontation with the data.(Mayer,L.S., 1980)
2.3 The differences in application between descriptive, exploratory, and confirmatory analysis methods of business and economic data.
Create a new hypothesis Briefly describe a given data set
Check the validity of the given hypothesis Discover new knowledge Based on quantitative data Based on existing research
There weren't any specific theories before
Based on available data set to perform description
Based on previously studied hypotheses
Describe or summarize the characteristics of a sample or data set
Make the results more meaningful
The overall design has the flexibility
The overall design must be tough
The overall design has the flexibility
Table5: Thedifferences betweenmethodsof descriptive,exploratory,and confirmatoryanalysis Example: Application in my research
Evaluate the differences in application between descriptive statistics,
Figure15: Thedifferences betweendescriptivestatistics andinferential statistics
In my survey, I have made a quantitative description of the factors that assess Biti's brand awareness in Da Nang.
I have done the description of the properties, it uses measures with central propensity i.e mean, median, method, and measures of dispersion i.e range,standard deviation, degree quartile deviation, and variance, etc The data is
- 33 - summarized in a useful way by me, with the help of numerical and graphical tools such as charts and graphs to represent the collected data accurately.
Reference statistics are all about sample-to-population aggregation, i.e analysis results of a sample can be inferred to the larger population from which the sample was taken For the census population in my survey, the study population is 1.2 million people in Da Nang However, when it is observed that, it is very difficult to survey all 1.2 million people because the number is too large Therefore, I have selected 100 samples i.e 100 people out of 1.2 million to conduct the survey. Measuring association
Any different factor or coefficient used to quantify the association between two or more variables is considered a measure of association in statistics Linkage measures are used in many different areas of research Any of a number of different studies, such as correlation and regression analysis, can be used to establish the association measure (Although the terms correlation and association are frequently used interchangeably, correlation refers to linear correlation in a stricter sense, whereas association refers to any link between variables) (Haug, M Gerard , 2019)
The method used to determine the strength of an association depends on the characteristics of the data for each variable Data may be measured on an interval/ratio scale, an ordinal/rank scale, or a nominal/categorical scale These three characteristics can be thought of as continuous, integer, and qualitative categories, respectively (Haug, M Gerard , 2019)
In my survey, I applied the Spearman rank-order correlation coefficient to design measurements for the survey.
In my case, I have 7 factors to help identify Biti's brand in Da Nang I want to determine the hierarchy of these factors to effectively deploy future development strategies I have ranked the factors on that Likert scale ( from 1 to 5 ) where 1 is the lowest rating Spearman is implemented to measure the association between ratings.
From there, it shows the strength of the association between the variables measured on the rating scale The data is the result of ranking factors.
Conclusion
From my analysis, I have clarified the nature and process of business data/information from different sources The definitions of data, information, and knowledge have been given by me How data can be turned into information and information into knowledge has been analyzed through three examples In addition,
I analyzed and evaluated data from different sources through three methods of analysis Descriptive, exploratory, and confirmatory I have outlined the strengths and weaknesses of each of these methods Also, highlight the difference in the application of these three methods Next, I analyzed the qualitative and quantitative data and gave illustrative examples I was also able to distinguish between population and sample milk Furthermore, I have performed descriptive statistics for the data set that I have collected through the survey table Finally, I evaluated the differences in the application of descriptive statistics, inferential statistics, and measuring association I hope, through this report, will provide the necessary information for the company's managers to develop a plan to improve decision making and information management through statistical methods the most effective.