Analytical Decision Making Assignment
The descriptive analysis answers the question of what happened (Duan and Xiong, 2015). For example, Econet will seek to determine how many people bought airtime last month; the retail department- the average volume of sales at weekends; etc. Another example of Econet it could choose target product categories based on income analysis, monthly profit by product group, revenue by product group, total quality of metal parts produced by month. The descriptive analyst is introductory, retrospective and answered the question “What happened?” At present, about 80% of the company’s analyzes are analyzed, which is why it analyzes the most common type. Descriptive analytical controls of raw data from multiple data sources to provide valuable information about the past. The results, however, are simply what says that something is wrong or is not right, without explaining why. As a result, highly data-driven companies, such as Econet, do not subscribe to descriptive analysis alone and do not combine it with other types of analysis.
Descriptive analyzes are usually visualized in simple reports, dashboards and scorecards with data visualization software. It is good to praise this information performance by using business intelligence tools. For data analysis to provide the current value, a problem must first be defined. This is perhaps the most important aspect of descriptive analysis and how it opens the door to broader solutions. This, however, clearly indicates its limitations. While other analysts are more thorough about specific issues and re-enact next steps to becoming a business, descriptive analysts simply define the problem (Yaqoob et al., 2019). For many companies that consider their descriptive analysis “data-driven” is only the first of many steps.
Diagnostic analytes are also retrospective, but “why” looks for the problem encountered in the descriptive analysis. In this phase, historical data on other data can be evaluated to answer the question of what happened. Diagnostic tests make it possible to get drunk, find treaties and identify patterns (Nayebi et al., 2015). Diagnostic analysis is diagnosed in companies when it provides a thorough overview of a particular problem. At the same time, a company must have detailed information about its layout, otherwise the datasets can be disabled individually for each problem and each hour of use. Diagnostic analytic analysis is the essential “next step” to something as a descriptive analysis. Diagnostics analysts also provide historical data from a company on many internal sources. This analysis is more complex and reveals that analytical-scale data provide models, trends, and correlations. This can use data mining techniques such as regression analysis, anomaly detection, clustering analysis, and so on.
The largest set of diagnostic analyzes is able to contextualize a business problem through a number of data models. Although diagnostic analyzes conclude at the speed and performance of machines, it is important that human analysts do not misunderstand musical means such as the “cash-in” of a business problem. Instead, this information should be used to promote support. For many companies, “what” is the problem and “why”, it can prevent enough, but for some, the future offers more valuable answers. This is where predatory analysis comes in.
Predictive analysis indicates what is likely to be appropriate. It uses the conclusions of descriptive and diagnostic analysts to study trends, groups and exceptions, and to identify future trends, making it a valuable tool of prestige (Yaqoob et al., 2019). Despite the many benefits provided by the forecasting analysts, it is essential to understand that performance is only an estimate,that management depends on the quality of the data and the stability of the situation, so that it is treated with caution and that it is improved. For example, thanks to predictive analytics and the proactive approach, Econet, for example, introduces call, sms and data bundles likely to cost less, and sets up marketing activities to promote them; A management team can assess the risks of introducing such products on their business based on analysis and performance.
Predictive analysis, unlike the two previous analyzes, is forward-looking and somewhat proactive with its results. The attempt to talk about what will probably come out of it and is one of the things considered as “advanced analytics”. This analysis is very complete and made possible by advanced technologies such as machine learning, data mining and predictive modeling. With the targeted data and algorithm, companies must use the same error twice with predictive analytics. The public service is also a lot of industry. For a company like econet that operates in the telecommunications and mobile banking sectors, this is very useful. Predictive analytics, although many of them, can cause some breakdowns. First, it is important to understand that erroneous data always leads to a monetary analysis. Predictive models built with insufficient information will only create more threats for a company. In addition, the models must regulate the silence, duplication and refinement of data analysts and scientists to ensure that they generate the right results (Duan and Xiong, 2015).
The prescriptive analysis of the data aims to describe literally the action to be taken to prevent a future problem or to take full advantage of a promotional trend. An example of a prescriptive analysis from our business study is that Econet is able to identify customer retention opportunities based on customer analytics and sales history. This type of state-of-the-art data analysis uses not only historical data, but also external information via the nature of statistical algorithms. In addition, the normative analysis tools of technology and technologies, such as machine learning, business rules and algorithms, facilitate execution and management. Therefore, before deciding to create analysts specialized in the writing of newspaper articles, one must compare a mandatory effort required to an expected added value (Yaqoob et al., 2019).
With different types of analysis, companies can choose for free the importance of their depth to delve into data analysis to best satisfy their business students. While descriptive and diagnostic analysts offer a reactive approach, and users are proactive. At the same time, current trends show that more and more companies are analyzing and selecting large data. This analysis is extremely complex and requires a data scientist or scientist with priority knowledge of press releases. Press analysts typically use heavy machine learning to check and identify new rules. Analysts in charge of writing for the press face another challenge: their barrier to entry for many companies. This type of analysis can be expensive to generate and seek the help of data specialists – an area of abundant questions. Written press analysis is not a so-called current classification, but if data science becomes more common, we will see more affordable publishing options.
Primary Research Methods
According to Bernard (2017), primary data is all data that the author has received “first-hand” from its original source as part of the “adaptation” component of its research. There is no data previously collected by anyone. As a result, primary data sources are included: individual research in the form of evaluation, interviews, questionnaires, interviews, etc. After the end of the secondary research, the processes sought study and deepen the identified problem. During this phase, the researcher uses the main methods of data collection to collect research data using various methods such as questioning, assessments, conversations, surveys and focus groups, and so on. It is very important that the data is collected by a primary collection method not used by third parties and open to everyone, unless they have the rights and permissions of the authorities.
Interviews offer marketing researchers the opportunity to deepen their thoughts on topics of interest to those who want to better understand them (Brannen, 2017). Research projects that use this method usually have a relatively small number of these investigators and they realize the exact characteristics of the target audiences that researchers want to understand. For example, a pharmaceutical company may want to understand medical documentation when determining which drugs prescribe certain medical conditions. An enterprise software publisher can discuss with a “powerful user” of the product the limitations they see in the current product and some improvements they would like to see.
The interviews are structured in a discussion tool. The interviewer asks questions and then hears the answers correctly. He sometimes asks lists of questions for clarity and insight. In-depth interviews provide an opportunity to put light under the surface and are problematic for more reliable answers and more recent responses to interviewer questions. Often these interviews help researchers with the questions and answers they need to include in a quantitative survey (with more participants). Clear conversations can also be combined with behavioral concepts to better understand why people do what they do: “What do you think when …?” Or “Why did you do that …?” A long interview is an important message for an intervention interview. It is difficult to speak to people who have more than thirty minutes of conversation, so the discussion guide and interviewer should focus on covering time-based scripts.
Focus groups are numerous, as are full-fledged investigators, except that they place small groups (usually 6 to 12 people) earlier than one person at a time. As with interlocutory conversations, focus groups attempt to delve deeper into topics of interest with individuals who wish to better understand the perspectives of researchers. Focus groups have the additional benefit of inviting their peers to learn about the topics covered in the question, so that researchers do not only hear from one individual, but also listen to the group’s interactions (Brannen, 2017).
Bernard (2017) states that research and questionnaires are another wonderful and very effective way to conduct basic brand research. The term “judgments” is a broad term that describes many problems, such as mass surveys, navigation forms, research interviews, and customer satisfaction. One of the most common examples of this search method is the customer feedback form when billing at a restaurant. This is a simple method of knowing if the customer is satisfied with the operation of existing products and services or changes that they would like to see. The recipes are also provided in the form of web questionnaires which, every day, allow the company to collect a lot of comments and analyze them for later administration.
The proliferation of social media provides an excellent opportunity to know precisely which key people say in what type of message acquired by marketing. Social listening is a systematic process to follow what is said on a topic described in forums such as Facebook, Twitter, LinkedIn, blogs and even mainstream media. When they are listening, researchers see both positive and negative perspectives. Social listening helps markers not only those who say what, but also who influences who should help them. Social listening can be passive, with markers that consist mainly of following trend topics and regulatory rules for these topics. Social listening can also be done in a more direct and proactive way by asking or referring questions to a target group – a set of bloggers and influencers or a social media community – and, for example, “tell me what What do you think”.
Secondary research methods
According to Brannen (2017) secondary data indicates that the author is not responsible for collecting first-hand data. It then collects all the data from each other and presents them in various forms: magazines, reports, archives, annual reports of companies, newspapers and magazines, conference equipment, Internet, books, etc. At the time of collection, the researcher already uses available research data in the form of research papers, books, journals, unpublished research, journals, public websites and research articles, etc. The research has already been done by the researcher on official research and the researcher is allowed to use this material in her research with sufficient references to the sources to pose more research problems that she proposes. These secondary tools help the researcher to learn more about the research theme as well as the work and experiences of each researcher to advance research or research on the topic of research. Finally, you will find these questions and an additional contribution to the body of knowledge.
Internal sources are those kinds of secondary market research sources that already exist and are collected in the business’s database or file system. Internal sources include information that has already been collected by the company and proves useful for future projects, etc. For most businesses, internal sources may prove enough to develop new products and services, and this may not require them to look outside.
Typical descriptive statistics (also descriptive analysis) are the first level of analysis. This helps researchers understand data and capabilities (Bernard, 2017). Some descriptive statistics used are:
• Languages: numeric average of a set of values.
Median: middle of a set of numeric values.
• Mode: The most common value between a set of values.
• Percentage: used to express the relationship between a value or group of respondents in the data and a larger group of respondents.
• Frequency: The number of times has been found.
• Range: the highest and lowest value of a set of values.
The description of the statistics gives absolute numbers. However, they do not explain the principles or reasons that motivated these figures. To apply descriptive statistics, it is important to determine which one is best for your research question and what you want to show. For example, a percentage is a good way to see the distribution of respondents’ communities. Descriptive statistics are particularly useful when research is limited to the actual sample and should not be transferred only to a larger population. For example, if you compare the percentage of children in two different villages, descriptive statistics are sufficient. Since descriptive analysis is mainly used to analyze only variables, we often talk about univariate analysis.
As researchers collect data on a problem from their population, they modify the results for the entire population or target group. Inferential statistics are used to compare results and give examples of a larger population.
These are complex analyzes that show the relationship between several different variables, rather than a single variant. They are used when the researcher has to look for absolute values and understand the relationships between the variables.
Some types of infinite analysis are:
Correlation: This describes the relationship between two variables. If a correlation is found, it means that there is a relationship between the variables. For example, higher people may have a higher weight. As a result, size and weight are correlated with each other. This does not mean, however, that one variable changes in the other direction (because weight gain prevents people from growing up).
Regression: This shows the relationship between two variables. For example, regression can help us gain weight.
Analysis: This is a statistical procedure used to test the extent to which two or more groups differ or differ in an experiment. In most experiments, there is a big difference in the differences between the search results. For example, to understand the relationship between the number of children in a household and socio-economic status, a researcher can include a family collection of any socio-economic status and ask what is the ideal number of children . Variant analysis will be used to check whether the difference between group responses is statistically significant or random.
The choice of inferential statistics depends entirely on the purpose of the research. As in descriptive statistics, it is best to identify the appropriate survey statistics for your research questions. Since inference statistics are used to determine the relationship between two or more variables, they are called bivariate analysis (limited to two variables) or multivariate analysis (if there are more than two variables).
Qualitative Data Analysis Methods
Various methods are available to analyze qualitative data. The most commonly used data analysis methods are:
• Content Analysis: This is one of the most common methods of qualitative data analysis. It is used to analyze documented information in the form of texts, media or even physical objects. When using this method depends on the research questions. Content analysis is mainly used to analyze conversation parsers.
• Narrative Analysis: This method is used to examine the content of various sources, such as interviews with respondents, field observations or research. It focuses on the use of stories and experiences shared by people to answer the research question.
• Discourse analysis: like narrative analysis, discussion is used to analyze people. However, it focuses on the analysis of the social context in which the communication between the researcher and the respondent takes place. Discourse also analyzes the respondent’s day and uses this information in the analysis.
• Ingrained theory: it refers to the use of qualitative data to explain the cause of a certain phenomenon. To do this, he studies various similar cases in different contexts and uses the data to receive business explanations. Researchers can modify statements or create new ones by studying more cases, until they result in a statement that is appropriate in all cases.
These methods are the most common. However, other methods of data analysis, such as Conversation Search, are also available (Brannen, 2017).
Today, organizations are increasingly influenced by the growing uncertainty of the macroeconomic environment. In recent years, many organizations have experienced a recession or have fallen into crisis due to some external factors. These companies have one thing in the community, not quite predictable with the future. And now, a powerful tool widely used by companies to solve this problem. It is a tool organizations can think about in the future, starting by looking for new for-profit trends in an uncertain future, helping decision makers to help find mental models and find the right direction for the future. to come up. (Raford, 2015)). Social planning is a strategic decision-making instrument that does not focus on a sufficiently prestigious future, but creates a number of possible preferences, whether reliable or uncertain (Bowman, 2016). There is no doubt that companies can simply benefit from scenarios, and here are the different powers of scenario plans.
The use of the scenario planning manager may have a system badge. Since sugar shooters always play a clear role and think of a variety of possible scenarios in the next steps, Scenario planning can be a widespread mechanism before problems arise, managing the possible scenarios based on a series of scenarios. Logical and empirical facts, corporate culture all lead to trends in the evolution of operations, shape and influence of structure change in structure. (Chermack, 2017). In addition, scenario-based administrators may think of systems that take into account the various factors that they feel can influence good common analysis and avoid narrow personal biases. Using scenario plans, contractors can anticipate possible future scenarios so managers can be more aware of the external environment and status and future precipitation, so that they can provide internal resources for an external environment more effectively and rationally.
Systems planning helps us improve by this morning. It is good to help organizations deal with future uncertainties by becoming aware of weak signals and making them better able to cope with new situations in order to live and thrive. A scenario plan can be defined as a business plan aimed at targeting, correcting, reducing risks, reducing delays and maximizing profits. Although the objectives of the organizations adopting the scenario plans are the same, the implementation of Scenario Planning for the first time poses many problems.
The company will mainly face the following problems (Bowman, 2016; Chermack, 2017):
The costs of scenario plans in fact depend on many variables such as the size of the organization, the duration of the scenarios, the teams and partnerships in the strategic planning process, and the methods of analysis and data collection involved in these scenarios. the planning process.
To create accurate scenario performance, there is a lot of temporal noise. Typically, 6 to 12 month plans may require multi-scenario plans for some organizations in depth. Donations and information from various sources must be collected and interpreted, making scarification even faster.
Scenario planning must use several means, besides the money, suggesting to form a scenario management team that asks external experts to collaborate with industry, which is also due to the fact that it also publishes the data information on the company as it utilizes several departments.
For a company, it is not necessary to say that more scenarios are the best future for society and, of course, make no sense for business development if the company has too little contingency planning. All scenario plans must establish management plans and appropriate measures. Too many scenarios do not require time and money, but also increase the company’s operating costs. However, as some scenario plans do not allow society to talk about the future. to prevent
Duan, L. and Xiong, Y., 2015. Big data analytics and business analytics. Journal of Management Analytics, 2(1), pp.1-21.
Yaqoob, I., Salah, K., Imran, M., Jayaraman, P.P. and Perera, C., 2019. The Role of Big Data Analytics in Industrial Internet of Things. arXiv preprint arXiv:1904.05556.
Nayebi, M., Ruhe, G., Mota, R.C. and Mufti, M., 2015, November. Analytics for Software Project Management–Where are We and Where do We Go?. In 2015 30th IEEE/ACM International Conference on Automated Software Engineering Workshop (ASEW) (pp. 18-21). IEEE.
Bell, E., Bryman, A. and Harley, B., 2018. Business research methods. Oxford university press.
Bernard, H.R., 2017. Research methods in anthropology: Qualitative and quantitative approaches. Rowman & Littlefield.
Brannen, J., 2017. Mixing methods: Qualitative and quantitative research. Routledge.
Quinlan, C., Babin, B., Carr, J. and Griffin, M., 2019. Business research methods. South Western Cengage.
Raford, N., 2015. Online foresight platforms: evidence for their impact on scenario planning & strategic foresight. Technological Forecasting and Social Change, 97, pp.65-76.
Bowman, G., 2016. The practice of scenario planning: An analysis of inter‐and intra‐organizational strategizing. British Journal of Management, 27(1), pp.77-96.
Chermack, T.J., 2017. Foundations of scenario planning: The story of Pierre Wack. Routledge.