Statistical Analysis of Complex Problem-Solving Process Data: An Event History Analysis Approach
Yunxiao Chen 1 * , Xiaoou Li 2 , Jingchen Liu 3 and Zhiliang Ying 3
Complex problem-solving (CPS) ability has been recognized as a central 21st century skill. Individuals’ processes of solving crucial complex problems may contain substantial information about their CPS ability. In this paper, we consider the prediction of duration and final outcome (i.e., success/failure) of solving a complex problem during task completion process, by making use of process data recorded in computer log files. Solving this problem may help answer questions like “how much information about an individual’s CPS ability is contained in the process data?,” “what CPS patterns will yield a higher chance of success?,” and “what CPS patterns predict the remaining time for task completion?” We propose an event history analysis model for this prediction problem. The trained prediction model may provide us a better understanding of individuals’ problem-solving patterns, which may eventually lead to a good design of automated interventions (e.g., providing hints) for the training of CPS ability. A real data example from the 2012 Programme for International Student Assessment (PISA) is provided for illustration.
About Data Analytics
Data Analytics is highly quintessential for understanding what things should be done and which ones must be avoided at all costs. Basically, Data Analytics is the extraction of valuable information from unstructured or structured data by using various algorithms, tools, methods, and techniques. This technology also helps to display all the things that are efficiently working on the various platforms of social media, marketing, content writing, and much more. Data Analytics particularly helps to discover the likes and dislikes of your customers which you can take into account for the betterment of the whole organization.
Most organizations which have not much awareness regarding the latest technologies are bound to use the old and lesser advanced versions of various technologies. What they fail to understand is that these kinds of technologies are only going to make the situations worse and much more complex for any IT professional to handle. That is why it is much better to use Data Analytics to solve a large number of problems.
This is because it contains advanced tools that can be used to bring about greater change in lesser time. About 70-80% of the time is generally used to engender various files for analysis and then the rest of the time is actually used to solve the given problem. The tools of Data Analytics undertake two major criteria to resolve various sets of problems of a given organization and these are:
The rapid and increasing demand for such advanced tools of Data Analytics has forced various entrepreneurs to shift their services to such technology. This technology enables organizations to obtain answers to even those questions which they found very complicated to answer. The unseen wonders of Data Analytics will be put into action and the organizations will no longer need the services of the rest of the programming technologies and support.
Enroll in our data analytics bootcamp program offered in collaboration with the top universities in the US. Connect with our experts to learn more about program details, curriculum, payment structure, learning modalities, job assistance and other value propositions offered.
Problem solving with data
Every model of problem solving emphasizes the importance of information, knowing as much about the problem as possible: The history of the problem, the causes and origin of the problem, previous solutions that worked or failed, the scope of the problem, the impact of the problem.
Assembling data about the problem often begins with input from stakeholders, including employees, volunteers, customers, members, or the general public. Initially, the information may come from a suggestion box, a complaint, a notice from a landlord or a government agency, a survey, a quarterly or annual report, a newspaper or mass media story, a new book, or an academic article. Or it may come from an expert or a consultant.
A quick example from the news is the hacking of credit card records at Target. This could pose a problem for any business that accepts plastic — which would be every business. The Internet can be an especially useful source for data on such current event issues.
The sba.gov website, for example, includes an article on security that addresses the computer security problem, as well as several related problems. The article, by Caron Beesley, a small business owner, writer, and marketing communications consultant, begins as follows [italics added]: “How secure are your small business assets from fraud, identity theft and cybercrime? According to the Association of Certified Fraud Examiners (ACFE), companies with less than 100 employees lose approximately $155,000 as a result of fraud each year. Small businesses also have a higher fraud rate than larger companies and non-business owners. One of the most frequent sources of fraud is credit card abuse – largely due to the fact that few business owners actually take the time to go through every line item on their bill or choose to mingle business and personal accounts. Other sources of fraud stem from an overall lack of security across the business – such as inadequate network and computer security and a lack of background checks when hiring employees.”
Raw data is rarely useful in problem solving without analysis. In the example just cited regarding fraud against small businesses, annual losses are given as $155,000. But what does this number mean? How was this figure determined? Surveys? Tax records? Estimates based on prior years? Overall trends affecting all businesses large and small?
Obviously, the small business owner wants to know how serious this problem is for his business. That figure alone, without proper analysis, does not help him or her decide whether to address credit card fraud as a major problem requiring what might be expensive solutions (a new computer system or new software?).
Now more than ever protecting revenue is important to your business. Retailers are losing over $100 billion in fraud losses each year, mostly due to identify theft and charge backs. To help effectively battle the fraud problem burdening U.S. retail merchants, LexisNexis® Risk Solutions teamed up with Javelin Strategy & Research to conduct a landmark study on retail fraud.This annual study comprehensively assesses the cost of criminal methods, industry solutions, and merchant needs.”
After considering the accuracy of the data, there are other issues that arise. Is the data relevant to the problem at hand? In our example, the total number of Visa cards in circulation, or the fees charged by the credit card companies, or the average purchases in the holiday shopping season are useful pieces of data for business planning. But they are not applicable to the specific problem of fraudulent use of credit cards at small businesses.
Discussion: “Many heads are better than one,” is one of those truisms that is actually true. Regardless of your situation — whether you have a personal problem to solve, or are in a formal task force charged with problem solving, different perspectives are not only useful, they are essential. Everyone knows about his or her own blind spots, prejudices, automatic habits, hidden (even unconscious) agendas, and group think. These can lead one astray. But the solution to this problem is pretty simple. Get input from others with differing perspectives and knowledge and points of view. They will help balance all the ideas that may be distorted by your personal thought process.
Negotiation becomes necessary in any group problem-solving process where consensus is not reached quickly and easily. Consensus is not often easy and is always time consuming and energy draining. The answers to “What exactly is the problem we are trying to solve?” will vary dramatically from person to person. It is common sense that the salesman and the comptroller do not look at a drop in quarterly sales the same way.
At last, the problem solvers have a workable, practical definition of the problem that they can use going forward. We are finally going to tackle absenteeism or competition in the market or new technology in our industry.This is where the committee or the task force is going to focus its attention and resources.
Sources:
https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00486/full
https://www.quickstart.com/blog/why-data-analytics-is-important-in-problem-solving/
https://www.universalclass.com/articles/business/problem-solving-strategies/problem-solving-strategies-research/gathering-data.htm