Statistics: Managerial Report Paper.

#137461710

Topic:

Rerunning SPSS

Discipline:

Statistics

Academic level:

Undergraduate

Category:

Analysis by SPSS

Software:

SPSS

Deadline: 18hrs

(Instruction)

1.The attached file of ‘19-0330 outputs.spv’ was produced by SPSS based on the database shown in the file of ‘CSR firm ranking 2017 top 300 in CSR White Paper 2017, English version.’ However, ‘19-0330 outputs.spv’ is missing all ‘log’ for each analysis.

Need to show all ‘log’ each analysis on the outputs.spv by rerunning SPSS based on the same database.

2. Need to provide a justification to use chi-square test for the test of association in the case of this database.

3. For your information, chi-square tests include the following pairs.

(1) Chi-square test for variables finance_total and HR utilization.

(2) Chi-square test for variables finance_total and environment.

(3) Chi-square test for variables finance_total and corporate governance.

(4) Chi-square test for variables finance_total and societal.

4. For your information , regression analysis include the following.

(1) Regression analysis where finance_total is dependent variable

(2) Regression analysis where profitability is dependent variable

(3) Regression analysis where stability is dependent variable

(4) Regression analysis where firm_size is dependent variable

5. For your information, dependent variables include four variables namely: HR utilization, environment, corporate governance, and societal, while independent variables include other four variables namely: profitability, stability, firm size, and finance_total which is a sum of profitability, stability, and firm _size.

(Attached files)

1) 19-0330 outputs.spv (IBM Statistics Viewer)

2) CSR firm ranking 2017 top 300 in CSR White Paper 2017, English version

**Managerial Report**

- Basic analysis from excel; graphics and numerical summaries

Descriptive statistics for percent under 21 and fatal accidents per 1000

PercentUnder 21 | Fatal Accidentsper 1000 | |

Mean | 12.26190476 | 1.922404762 |

Standard Error | 0.483237634 | 0.165257284 |

Median | 12 | 1.881 |

Mode | 8 | #N/A |

Standard Deviation | 3.1317378 | 1.070989605 |

Sample Variance | 9.807781649 | 1.147018735 |

Kurtosis | -1.137109498 | -0.974888754 |

Skewness | 0.210357273 | 0.193164404 |

Range | 10 | 4.061 |

Minimum | 8 | 0.039 |

Maximum | 18 | 4.1 |

Sum | 515 | 80.741 |

Count | 42 | 42 |

Scatter plot

- Regression analysis; to assess the association between the two variables
- What is your independent variable, X, in this problem?

Age (Percent under 21)

- The response variable y in the scenario

Number of fatal accidents per 1000

- Regression line for scenario

y = -1.5974 + 0.2871x

- Slope value from the equation and its meaning based the scenario

Slope = 0.2871

When the X value changes by 1% then the number of fatal accidents increases by 0.2871

- y intercept and its meaning based the scenario

y-intercept = -1.5974

it is the number of fatal accidents when the percent under 21 is zero.

- Coefficient of determination R
^{2}and its meaning based the scenario

R-squared = 0.704571 = 70%

The variations in the number of fatal accidents is explained by X = percent under 21. The data is 70% close to the regression line.

- Hypothesis testing for the slope alpha = 0.05.

Null Hypothesis, H_{0}: The percent of licensed drivers aged below 21 years is not linearly related the number of fatal accidents per 1000 accidents in the 42 cities

Alternative Hypothesis, H_{1}: The percent of licensed drivers aged below 21 years is linearly related to the number of fatal accidents per 1000 accidents in the 42 cities

Level of significance; alpha = 0.05

Test statistic from regression analysis; t stat = -4.29792

Respective p value = 0.000107

Decision; the p value is less than the significance level: 0.00017<0.05 hence the null hypothesis is rejected

Conclusion: the null hypothesis was rejected; therefore, there is a significant relationship between the two and age (Percent under 21) is a significant predictor of the number of fatal accidents per 1000 in the sampled 42 cities.

- Prediction o the number fatal accidents given that the percent under 21 = 5%

y = -1.5974 + 0.2871x; x = 5

y = -1.5974 + 0.2871 * 5

= -0.5385

Therefore, a city with 5% of its licensed drivers under 21 years would expect zero fatal accidents per 1000 licenses

- Conclusion and recommendations about the two variables

The percentage of licensed drivers with age below 21 years in the 42 cities was found to be a significant predictor of the number of fatal accidents in the same cities. It was found out that the higher the percentage, the higher the number of fatal accidents. Therefore, I would recommend that U.S. Department of Transportation should issue less or no licenses to those under 21 years if the fatal accidents were to be minimized.