Management and Spurious Correlation - Research Paper Example

? Management and Spurious Correlation Introduction Management and spurious correlation can be described as a mathematical relationship whereby there are two events or variables that have no direct or casual connection with each other but still may be identified as two events or variables that have a connection due to a wrong identification or due to coincidence. A false assumption is whereby there are two variables that presumed to be correlated when in the real sense, they are not. They can also be confused to be related because of the presence of an unseen or confounding factor (Kirk, 2008). Spurious correlation or relationship is in most cases something that results from a third factor that is not visible or apparent at the time when the experiment or examination was done. The term spurious is derived from the Latin origin. It is meant to mean not true or false. Management and spurious correlations are terms that are used in the field of statistics to make assumptions about certain things and calculations. When looking at the term spurious relationship, it can duly be noted that it is commonly used in statistics and in a particular way to provide certain answers. It is specifically used in particular to help in the experimental research techniques which would go a long way in answering some questions. Using spurious correlation in experimenting in research techniques help in the understanding and prediction of other relationships, namely direct causal relationship. An example of a direct causal relationship can be (X > Y). There is also a relationship called a non-causal correlation. A non-causal correlation may and can be created spuriously by a factor that is called an antecedent. An antecedent, for example, causes both the relationship as shown; (W > X and W > Y). Certain variables, known as intervening variables if undetected may make the indirect causations seem to look direct (Wooldridge, 2009). An example of an intervening variable can be seen in the form of (X > W > Y). Because of the effect of indirect causations being made to look direct, correlations that have been identified through experiment are not seen to represent relationships otherwise known as casual. This is unless spurious relationships can be ruled out. Only then can it happen. The purpose of this essay is to bring to light the importance of management and spurious correlations as used in statistics. As the name suggests, the essence of this essay is to bring out the importance of identifying correlations before they become misleading. This topic is important because many people tend to confuse the relationship between variables and by the end of this essay, it will be clear on the methods. In order to become a statistician or a social scientist, it is paramount that one understands that correlation be shown by being proven statistically. The purpose of this paper is to illustrate the widespread occurrence of spurious correlation (Kleinbaum & Kleinbaum, 2008).  In order to test and prove whether a correlation between two variables or constants is genuine or spurious, there are some additional variables and equations that have to be introduced in order to widen the parameters of getting legit results. There have to be sufficient assumptions being made and they must be made in order to help with the proper identification. This proper identification is of the parameters of a wider system and will help in the obtaining of results. It the results are found that the two variables, which were original, are causally related in a wider system, then the conclusion of the correlation is that it is genuine. The difference between a true and spurious correlation is that a true correlation exists and does not have to be proven while a spurious one needs to undergo experiments in order to be proven to be wrong. Statistical research has been affected by the identification problem, where many statisticians have been unable to conclude on where the problem is. An example of spurious correlations can be made using the following model: Looking at statistics, the analysis of human resources when it comes to data usually involves the implementation and use of computer databases (Srivastava & Rego, 2008). These computer databases were initially made with the intent of processing transactions. As a course for their actions, computer databases are normally made to be administrators. Their main work revolves around the administration of records and also recordkeeping. Taking an example from this, those variables or aspects that are brought forward for analysis are not totally the ones that would be taken into account as the ideal set of variables or aspects given the main agenda of the analysis. One major setback is that there may be side effects of this analysis. One of the side effects is that when it comes to most cases, most of the critical variables that should be used in the analysis may be missing. When this happens, it is safe to say that the result of variables missing in the analysis would lead to results that would be called spurious results. The results of the analysis would be wrong, which might later lead to serious consequences. As a result of the analysis, the statistical data collected would bring about false information, information that was being relied on by another individual or organization. Suppose that one of the critical variables that are missing is correlated with color. A variable that relates with the critical variable will often correlate with color. These correlations are false mainly because their main factor in analysis is the critical variable that is missing. Spurious correlations are often at times used to show the aspect of discrimination when it comes to analysis in statistics (Bergh & Ketchen, 2009).  Spurious correlation is something that most people do not recognize. Most people ignore spurious correlation fallacy. When it occurs, the occurrence is very much real and in its own way is very persuasive. The sad thing is that even with its persuasiveness, it goes on unnoticed from the people. Many people have spoken and written about spurious correlation, and it has not yet proved its significance to the lives of the populace. Maybe part of the problem is that there has been the presence of various variations in terms of the definitions used by different authors. Different authors have shown and illustrated in different ways that the fallacy does certainly exist, described and discussed. The purpose of the way the authors have defined spurious correlation is not so as to pit against each other, but to bring the realization that it is a fact that continues to be ignored by many. The main agenda is to use the numerous repetitions in different books in order to create and awareness and to develop a conscience for the repetition so that the recognition of the false correlation becomes much easier (Bergh & Ketchen, 2009). A lurking Variable is the variable that can be defined as the one which brings about the spurious correlation. It can be defined as the third variable of which it is the cause of a correlation between the other two existing variables. When speaking about a lurking variable, it can be simply defined as the main source as to where a spurious correlation emerges. In experience with life, there are several things that are used to show the correlation and its use in real life. Examples have been made of aspects like the relation between cigarette smoking and school grades. Another study that has been done just to show the correlation between two different variables has been an experiment to show the relation between fat in the diet and the contraction of cancer. Others have come up with other experiments such as the one where a link exists between healthier babies and bottled water. Looking at the example on the relationship between smoking and school grades, it is safe to say that the issue about correlation has been addressed (Bergh & Ketchen, 2009).  An experiment was done once in a past study in order to determine whether they the researchers could identify the problem and solve it. The researchers went to a lot of extent and a great deal of trouble in order to test the theory that cigarette smokers were more than likely to have lower college grades in college that the rest, the nonsmokers. Consequently, in accordance to the aims of the experiment, it did. It came to be proven that there is a relation between the cigarette users and the grades of the students. The smokers had grades that reflected the rate of their health. Since cigarettes decrease the health status of an individual, so were the grades. They were decreasing significantly (Andersen & Taylor, 2008). On the other hand, those who did not smoke had perfectly good grades. This, according to statistical researchers, is proof that there is a relationship between two variables. Since then, this sought of examples has been shown to many. It is meant to show the general conclusion that in order to have good grades and be successful, then one has to give up smoking. In order to make this conclusion seem deeper, it is safe to say that one conclusive outcome of this experiment shows that smoking does damage the nervous system and therefore makes people dull than they already are. A study of this kind would mean that the facts are set straight. The correlation in the above experiment is one that no one can argue with. The correlation between the two variables is clear, brief and honestly portrays itself. The correlation shows very high significance. The lesson learnt is that an action often correlates with other actions. If an action C follows a certain action A, then it is safe to say that action A causes the action C. One unwarranted assumption that arises is that of being concluded that since taking cigarettes and low grades go together, then it is definitely smoking that causes low grades. But perhaps the researchers could be wrong. In another term, learners who acquire low grades do not take their studies seriously; or perhaps then because of the low grades, the students are just stressed out to the point that they turn to cigarettes (Kirk, 2008).  This realization can also be true, and based on the collected evidence; one can assume that the fault either lies on the cigarette, the student or the books. Looking at this from another prospective, it is true to say that there is a lot better deal at thinking this through than with all the facts discussed earlier. It is safe to assume that none of these things has led to the production of the other. But still, both are a product another factor; a third factor in the equation. From the assumption that there could be a student who does not take their studies more seriously assuming that it is because of the grades, they are forced to become tobacco addicts. There is another assumption; that because the researchers stated that tobacco causes dullness makes for there to be en established correlation between studies and smoking. The whole point to conclude here there are more explanations that people can choose from, but many prefer to ignore the many and concentrate on one (Andersen & Taylor, 2008).  Conclusion In the essay above, it has been observed that the essence of management and spurious correlation has been assumed by many. Many individuals make up their minds on the assumption to follow and ignore reading between the lines and making a clear judgment. It is safe to assume that statistically, the power of spurious correlation has been abused by many individuals, and they are not able to distinguish between the true correlation, and the false correlation. It is high time that people stopped making mistakes and look at the facts without leaving any option behind. References Andersen, M. L., & Taylor, H. F. (2008). Sociology: Understanding a diverse society. Belmont, CA: Wadsworth/Thomson Learning. Bergh, D. D., & Ketchen, D. J. (2009). Research Methodology in Strategy and Management, 5. Bradford: Emerald Group Pub. Kirk, R. E. (2008). Statistics: An introduction. Belmont, CA: Thomson/Wadsworth. Kleinbaum, D. G., & Kleinbaum, D. G. (2008). Applied regression analysis and other multivariable methods. Australia: Brooks/Cole. Srivastava, T. N., & Rego, S. (2008). Statistics for management. New Delhi: Tata McGraw-Hill Education. Wooldridge, J. M. (2009). Introductory econometrics: A modern approach. Mason, OH: South Western, Cengage Learning. Read More