It is undeniable that the success of any organization depends greatly on the organization’s labor force, otherwise referred to as human capital. Employees, by working for the company, are responsible for the growth and profit made by the organization. This paper addresses the issue of high turnover in an arbitrary manufacturing organization, Finally Paper Company by employing the relevant statistical research methods.
Finally Paper Company is a major paper supplier that imports office paper and supplies in large scale to clients all over California. Its effective performance is based on not only the availability of technical resources but also the competence of its staff in relation to the paper industry. This financial year alone, the company has lost 11 out of its 31 employees, an annual employee turnover rate of 43.1%.
With a reduced labor force, the same amount of work is left for people to handle who find it too much and end up not having a choice but doing a poor job. It also takes away experience and knowledge that cannot be monetarily measured. Causes of the high turnover in this organization are inadequate motivation among the workers, a fairly conducive environment and poor wages. Strong predictors for turnover include age, job satisfaction, tenure and individual perception of fairness.
The method selected to address turnover in Finley Company is descriptive logistic regression to come up with the relationship model between the independent variables (workplace data) and turnover, the dependent variable. We will utilize graphical and numerical methods to determine patterns in our data set summarize the information and present it in a meaningful form. Coupling that with inferential statistics, we will come up with a conclusion about the employee population based on the information provided.
The preference of this method over others is due to our need to represent the participant’s responses to our survey items so as to address the issue in question and determine the odds of a particular employee quitting. We use include averages and percentages and frequencies for ordinal and nominal data. After the analysis, we conduct inferential (parametric and non-parametric) statistics to draw conclusions on the significance of the relationship between our variables of interest. In our particular situation, we consider the employee turnover equation by asking whether or not they consider quitting their current job.
We also provide them with a set of satisfaction attributes to which they respond by giving their levels of agreement, a 1 for strongly disagree and 5 to indicate strongly agree. Sample appropriate attributes are: “The Company gives me the motivation I need to give my best at work”, “My workplace environment is conducive” and “My experience and qualification are considered for equal promotion of employees”. Additionally, we focus on other areas such as stress at the workplace, working hours, incentives and reward the company offers to encourage and appreciate hard work by the employees.
Taking the findings of the paper to be that the high turnover rate is caused by a blend of individual-specific factors, the employer needs to take action geared towards changing their employee’s decision to quit. They should offer attractive benefit packages and understand the needs and expectations of their labor force so as to design an appropriate employee retention program. In addition, they should implement a wage increase as much as their financial status can allow so as to give their workforce something to look forward to.
Linear regression is advantageous over other methods as it is intuitive and easy to use. Its diversity makes it fit in varying situations, ours included. It works well when the relationship between the response and covariates is is known to be linear. The shift of focus from statistical modeling to data analysis is a main advantage. The method is so prolific that it is incorporated into most statistical analysis software. It works well for predicting categorical outcomes quitting of employees during a specific time period.
A downside to logistic regression is that the output can sometimes be difficult to interpret. The method uses a set of independent variables to predict outcomes. In case the researcher uses or includes wrong variables, the model will have no predictive value at all. It requires all relevant independent variables to be identified beforehand. Another demerit of this method is that it is incapable of predicting continuous outcomes. Additionally, it requires that each data point being worked on be independent of all other data points. Lastly, logit models have the capability of oversimplifying situations leading to overconfidence. Models may appear to have exaggerated predictive power (UCLA, 2018).
- Bliss, B. (2000, February 28). The Business Cost And Impact Of Employee Turnover. Retrieved from https://www.ere.net/the-business-cost-and-impact-of-employee- turnover/
- Bernard, H. R., & Bernard, H. R. (2012). Social research methods: Qualitative and quantitative approaches. Sage.
- Robinson, N. (2018, April 5). The Disadvantages of Logistic Regression | Synonym. Retrieved from https://classroom.synonym.com/disadvantages-logistic-regression-8574447.html