Independent, Mediating, Moderating and Dependent Variables - Coursework Example

Contributing to Theory: Independent, Mediating, Moderating and Dependent Variables Contributing to Theory: Independent, Mediating, Moderating and Dependent Variables A mediating variable (mediator or intervening variable) is an intermediate construct that describes how effects or output will occur by justifying the relationship between the dependent and independent variables. It also accounts for the relationship between the criterion and the predictor. Mediator constructs explicate how the external physical events or circumstances take on the internal psychological significance (Springer, 2014). Unlike moderators, mediator variables explain why or how certain effects occurred. Intervening variables provide a causal link between variables, as well as define a relation. For instance, intervening variables explicate the statistical association between longevity needs and income because having money does not necessarily make someone live longer. However, the relationship between income and longevity needs might hold considering that people with higher income tend to have better medical care. A typical example of intervening variable is the medical care. It basically mediates or intervene the relation between longevity and income (Springer, 2014). All the correlations between predictor and mediator variables, mediator and criterion variables, as well as predictor and criterion variables are significant for general test for mediation. For instance, parents typically transmit their social statuses directly to their children. In the same way, they effectively transmit those statuses to their children through education. Parent’s status ➛ child’s education ➛ child’s status A mediating variable is vital for statistical analyses and research designs that aim to investigate how tobacco prevention programs impact or accomplish their effects on the use of tobacco. The theoretical approach to program evaluation and development mainly focuses on the action theory (Amine et al. 2012). In particular, the action theory explains how the tobacco prevention program would affect intervening variables. Similarly, a conceptual approach seeks to explicate how mediator variables are related to smoking. Tobacco prevention programs would ultimately obtain information on the intervening mechanisms; information would be very useful to the development of adequate programs (Amine et al. 2012). The information would further provide a valuable test of the theoretical or hypothetical basis of tobacco prevention efforts. Amine, Chakor and Alaoui (2012) investigate the influence of mediating variables on the theoretical analysis of ethics, relationship marketing and corporate performance. The goal of the study is to establish a conceptual model representing the different relationships (Amine et al. 2012). According to their findings, ethics occupied a predominant position in the management of the organization. Thus, the establishment of an ethical climate within the organization is an intangible resource with positive effects on the growth of performance. The results further indicated that relationship marketing acts internally in the enterprise through internal marketing. The ethics indirectly affects corporate performance across the four dimensions of internal marketing (relationship marketing) namely organizational structure, communication, organizational commitment, and job satisfaction (Amine et al. 2012). The four variables constitute mediating variables between ethics and performance. Potential mediating/intervening mechanisms include mediated effects through social norms, attitudes, and beliefs. Prior research indicates that changes in the social norms of society are a vital mediating mechanism for effective tobacco prevention (Amine et al. 2012). On the other hand, conceptual theory underlying tobacco prevention programs provide a basis or foundation for understanding the process in relation to mediating variables to tobacco use. For instance, cognitive factors, attitudes and beliefs, and experimentation process help to explain the etiology of smoking among adolescents (Amine et al. 2012). The mediator variables would then be modified to reduce tobacco use once the conceptual theory is in place. A moderator variable is a quantitative or qualitative variable that affect strength and direction of the relationship between the dependent (criterion) variables and independent (predictor) variables (Springer, 2014). It also affects correlation (zero-order correlation) between two other variables. Unlike mediator or intervening variables, a moderator variable specifies when certain effects will hold. In addition, it significantly moderates or influences the strength of the sort of relationship existing between two other variables. As a result, a moderating variable generates an interaction effect. Springer (2014) conducted a survey on the influence of mediating variable on the interdependence between job satisfaction and perceived difficulties in balancing work and family work. The study was conducted on a sample of 533 working people. 126 of the participants stated that they had problems with balancing work-family life while 269 of them said that the aforementioned problem was the least of their concerns (Springer, 2014). The results of the study show that the subjective work value and a person’s professional position were the key variables that moderated the influence or impact of perceived conflict on job satisfaction. Meanwhile, a dependent variable (measured or response variable) is the presumed effect or output in an experimental study. Typically, the values of a dependent variable depend upon those of the independent variable (Kaminski, 2007). In essence, an experimenter would measure the dependent variable in an experiment and determine how the variable is affected. In essence, the variable is referred to as “dependent” because it depends on or relies upon the independent variable. It is arguably impractical to have the dependent variable in a scientific experiment without an independent or independent variables. Kaminski (2007), in the article “Central Limit Theorem for Certain Classes of Dependent Random Variables”, mainly examines the Central Limit Theorem (CLT) for the dependent classes of variables. Kaminski computes the difference between characteristic function of the sum dependent random variables and the characteristic function of the sum of independent random variables of the same distribution (Kaminski, 2007). The resulting difference serves for defining the class/category of dependent variables. While dependent variable is often the presumed effect, an independent variable, on the other hand, is the presumed cause in an experimental or scientific study. An experimenter often changes or manipulates independent variables, which will in turn have a direct effect on the dependent variables (Kaminski, 2007). An independent variable provides the “input” which the model further modifies to change the “output”. For example, an experimenter may conduct a scientific investigation to establish or test the theory that vitamin supplements could extend life expectancy of an individual. In this regard, the amount of vitamin given to the participants or subjects during the experiment is the independent variable (Kaminski, 2007). On the other hand, life span is the variable (the dependent variable) being affected by the amount of vitamin given to the subjects (independent variable). In another experimental study involving a scientist who investigates the impact of a certain vaccine on 52 Ebola patients, the independent variable(s) was the administration of the vaccine – the timing and the dosage. The dependent variable, on the other hand, was the impact the vaccine has on Ebola patients. Lastly, an experiment that seeks to scrutinize a theoretical assumption that people with college degrees are more respectful of the law than those without degrees, the independent variable would be whether or not the subjects have college degrees (Jabbari, 2013). The dependent variable is the effect that college degrees have on how respectful the subjects are of the law. Lastly, Jabbari (2013) seeks to prove the sure limit theorems for weighted sums of the random variables. He obtains a version of the Glivenko–Cantelli lemma for pairwise NQD random variables under some fragile conditions, with {Xn,n≥1} be a sequence of random variables (r.v.’s), but may necessarily not be mutually independent (Jabbari, 2013). Many statisticians would wish to extend the aforementioned condition to virtually all dependent cases. However, one of the dependent structures is often negative dependence. References Amine, M. E. A., Chakor, A., & Alaoui, A. M. (2012). Ethics, relationship marketing and corporate performance: Theoretical analysis through the mediating variables. International Business Research, 5(8), 68-84. Becker, T. E. (2005). Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendations. Organizational Research Methods, 8(3), 274-289. Jabbari, H. (2013). On almost sure convergence for weighted sums of pair wise negatively quadrant dependent random variables. Statistical Papers, 54(3), 765-772. Kaminski, M. (2007). Central limit theorem for certain classes of dependent random variables. Theory of Probability and its Applications, 51(2), 335-8. Rubin, A., & Babbie, E. R. (2010). Essential research methods for social work. Belmont, CA: Brooks/Cole, Cengage Learning. Springer, A. M. (2014). Influence of mediating variables onto interdependence between perceived difficulties in balancing work and family life and job satisfaction. International Journal of Human Resource Studies, 4(3), 158-177. Read More