Answer

1. • • • Steps in Bivariate Analysis

         Having collected the required data, the next step in quantitative research is statistical analysis, which takes three forms depending on the number of variables in the study. The three forms or techniques are univariate, bivariate, and multivariate analyses. As the names suggest, univariate analysis is applicable when the data consists of only one variable, and thus no need for establishing cause or effect relationships; bivariate analysis looks at the relationship between two variables; whereas multivariate analysis technique is appropriate when the data set has more than two variables. Here, the focus is on bivariate analysis and the steps involved in statistical analysis using the technique.

         Steps in Bivariate Analysis

         As one of the simplest forms of statistical analysis, bivariate analysis is used to establish if there is an empirical or cause and effect relationship between two data sets (dependent and independent variables). According to Bertani, Di Paola, Russo and Tuzzolino (2018), bivariate analysis is more analytical and focus of comparing two data sets or exploring how the dependent variable is explained by the exploratory (independent variable). This way, the technique underpins the assessment of the dependence or cause and effect relationship between the outcome and the exploratory variable. Bivariate analysis is further classified into categories: dependence analysis and inter-dependence analysis. The other techniques for bivariate analysis in case the variables in question are quantitative variables are the simple linear regression and correlation analysis. No matter the type of analysis used, bivariate analysis typically follows four distinct but interlinked steps.

          

         Step 1: Define and Represent the Nature of the Relationship.

         The first step in analyzing bivariate data is to define and represent the association between the variables. As (Bertani, Di Paola, Russo and Tuzzolino, 2018) points out, this can be achieved by displaying the relationship graphically using a scatter plot. Also known as a dispersion diagram, a scatter plot shows visually how the quantitative variates co-variate either linearly or non-linearly. Although the diagram allows one to see the and pattern and shape of the relationship and visually identify how the independent variable values relate to the dependent variable values, it does not provide the means of measuring the intensity of the causal effect (Bertani et al., 2018).

         Step 2: Use Bivariate Scatterplots to Identify the Type and Direction of the Association

         Having determined the pattern and shape of the relationship, the next step is to identify whether the relationship takes a linear or non-linear, and thereafter identify if the direction is positive or negative. The association is considered linear if the point plotted on the scatter plot tend to follow a straight line, and non-linear if the points seem to follow a curved line. If the association takes a linear form, then it be further classified in terms of direction as either positive or negative (Zhang, 2016; Sims, 2000). In this regard, the association between variables will be considered positive if the line of the plot has a positive gradient. Specifically, the dots on the scatterplot go up as one moves from left to right of the plot.

         Step 3: Determine Statistical Significance of the Association

         This step involves using Chi-square test to establish if the association between the two variables is statistically significant; that is, if it differs from the null hypothesis, and is generalizable to the population.  

         Step 4: Measure the Strength of the Association

         Here, the degree to which the independent variable values explain the variation in the values of the dependent variable is identified. The first and perhaps the easiest way of determining the strength of the relationship is by measuring the amount or extent of scatter in the scatterplot, and classify the association’s strength as either strong, moderate or weak if the dots in the plot follow a single stream, moderate scatter, and high amount of scatter, respectfully. Other techniques include the correlation analysis, Lambda or Cramer’s V, and best line of fit (Palazzolo, 2018).

          

          

Answer

Assessing Risk and Reward

The Nissan Motor Company is one of the leading automobile makers in the world. The Japanese carmaker has primarily enjoyed a successful run, allowing it to enter various regional and international markets such as the United States. However, the changing business environment was not favorable to the company in 2019. Notably, the cooperation recorded losses amounting to 7.8%. The experience pushed the management into making tough decisions, requiring almost all of its North American workforce to go for unpaid leaves.

In late 2019, the company announced that the decline in sales necessitated a two-day unpaid leave for the North American workers. The stated days for the vacation were January 2 and 3^rd    (Chicago Tribune, 2019).  Notably, this move was a crucial decision for the company because of its conflicting impacts. Whereas on the positive side, it could help the firm minimize expenses, it threatened to affect the public perception of the company regarding employee welfare.

The rewards for the decision involved cutting expenses by not paying the workers on leave, which eventually would translate into reduced expenses. Another reward was that the decision could allow the company to optimize performance by evaluating employee performances then developing new milestones. However, on the low side, the company risked affecting its public image and brand name, especially in the North American market. As per Chakravorti (2010), the way an organization treats its employees influences the firm’s public perception. Thus, Nissan risked eliciting a negative public perception. With a distorted public image, the company could fail to revive its declining sales.

I would have advised the management of Nissan to utilize the Predictive Analytic perspective in determining the right decision to take. Ideally, the approach tries to predict what might happen in the future if particular decisions or actions are undertaken at the moment (Traymbak & Aggarwal, 2019). Looking at the situation at Nissan, the company needed to develop a goal such as increasing sales. After that, they would have made decisions aimed at realizing the set goal. In this regard, the predicted outcome could give the management an overview of whether more risks existed or significant rewards could be realized.

.