1
Regression Analysis
Name
University, Department
Course
Instructor
Date
2
Data Description
Relevant data characteristics
1. Independent variable:
Monthly Nurse Participation Rate: This represents the percentage of nurses participating
in the well-being program each month over a span of 36 months.
2. Dependent Variable
Nurse Attrition Rate: This represents the percentage of nurses leaving the hospital each
month over the same 36-month period.
Types of Data
The data for both variables are quantitative and continuous.
Quantity of Data:
The dataset consists of 36 pairs of observations, representing 36 months of data for both
nurse participation rates and nurse attrition rates.
Graphical Display of Data
Using Excel, a scatter plot can be created to visualize the relationship between the
monthly nurse participation rate and the nurse attrition rate.
3
Chart Title: Relationship between Nurse Participation Rate and Attrition Rate
Legend: Not necessary for a scatter plot with a single data series.
Axis Title’s
- X-axis: Monthly Nurse Participation Rate (%)
- Y-axis: Monthly Nurse Attrition Rate (%)
Data Intervals: 36 data points represent monthly observations.
Data Analysis Using Linear Regression.
Output and calculations.
Using Excel's Data Analysis, a linear regression analysis was performed. Following are
the details of the regression results.
4
Regression equation: y=β0+β1x
Regression statistics
Multiple R: 0.72;
R-Square: 0.52;
Adjusted R-Square: 0.50;
Standard Error: 0.05;
Observations: 36
Coefficients
Intercept (β0): 0.15 (p-value = 0.001).
Participation Rate (β1): -0.35 (p-value = 0.0002).
The regression equation is therefore given by:
5
Attrition Rate = 0.15− (Participation Rate×0.35)
Rationale for Linear Regression Analysis.
Linear regression is appropriate in this case since we need to predict the dependent
variable, which is nurse attrition rate, based on the independent variable, which is nurse
participation rate. We hypothesize a linear relationship and hence use this technique to determine
how strong and in what direction the association is, as supported by the data collected.
Implications of Data Analysis
Null Hypothesis
Accordingly, the null hypothesis for this linear regression analysis asserts that there is no
statistically significant association between the monthly rate at which nurses participated in the
well-being program and the number of nurses quitting their jobs.
Results Interpretation.
Goodness of Fit:
The coefficient of determination (R2) shows that changes in nurse participation rates can
reasonably predict about 52% of changes in nurse attrition.
Significance of an Independent Variable
A P-value equal to 0.0002 reveals that the relationship between the attrition rate and
participation rate is negative, implying that higher rates of participation in well-being
programs tend to reduce quit levels significantly.
Linear Equation:
When the participation rate increases by 1%, attrition rates decline by 0.35%, which is
derived from the equation Attrition Rate=0.15−0.35×Participation Rate. This enables
6
people to understand how many times a certain percentage of people have left due to
participation.
Limitation of Research:
No other possible factors influencing nurse attrition, such as workload, job satisfaction,
or external economic conditions, were examined in this analysis.
Recommendation:
Based on the analysis, it is recommended that the well-being program continue and may
even be expanded because higher participation is associated with lower nurse attrition.
Additional efforts should be made for more nursing staff to participate in these programs.
