Final Statistical Analysis Report

Introduction

This report examines workplace productivity, job satisfaction, burnout, and training effectiveness using four statistical analyses. Dataset 1 uses a between-subjects design, while Dataset 2 uses a within-subjects design. The analyses include a Chi-Square Test of Independence, Pearson’s Correlation, Independent Samples t-Test, and Paired Samples t-Test.

Analysis 1: Chi-Square Test of Independence Research Question Is there an association between employees’ work mode (remote or in-office) and their level of burnout?

Variables

Variable 1: Work Mode (Remote, In-Office) Variable 2: Burnout Level (Categorical)

Hypotheses

Null Hypothesis (H₀) There is no relationship between work mode and burnout level.

Alternative Hypothesis (H₁) There is a relationship between work mode and burnout level.

Statistical Test

A Chi-Square Test of Independence was conducted to determine whether work mode and burnout level are associated.

Results

The Chi-Square Test produced a chi-square statistic of χ²(2) = 13.27 and p-value p < 0.001.

Interpretation

This result suggests that work mode is related to employee burnout levels. In practical terms, employees working remotely and employees working in-office did show different patterns of burnout.

Analysis 2: Pearson’s Correlation Research Question Is there a relationship between job satisfaction and productivity score?

Variables

Variable 1: Job Satisfaction (Continuous) Variable 2: Productivity Score (Continuous)

Hypotheses

Null Hypothesis (H₀) There is no correlation between job satisfaction and productivity score.

Alternative Hypothesis (H₁) There is a significant correlation between job satisfaction and productivity score.

Statistical Test

A Pearson Product-Moment Correlation was conducted to determine the strength and direction of the relationship between job satisfaction and productivity.

Results

The analysis produced a correlation coefficient of r(28) = .99 and p-value p < 0.001.

Interpretation

This finding suggests that employees with higher job satisfaction tend to have higher productivity scores. The strength of the relationship was strong.

Analysis 3: Independent Samples t-Test Research Question Do productivity scores differ between employees working remotely and employees working in-office?

Variables

Independent Variable Work Mode (Remote vs. In-Office) Dependent Variable Productivity Score

Hypotheses

Null Hypothesis (H₀) There is no difference in productivity scores between remote and in-office employees.

Alternative Hypothesis (H₁) There is a difference in productivity scores between remote and in-office employees.

Statistical Test

An Independent Samples t-Test was conducted to compare productivity scores across work modes.

Results

The analysis produced a t-statistic of t(28) = -8.14 and p-value p < 0.001.

Interpretation

This result indicates that employees working remotely performed differently from employees working in-office in terms of productivity.

Analysis 4: Paired Samples t-Test Research Question Does employee productivity change after training?

Variables

Independent Variable Time (Before Training vs. After Training) Dependent Variable Productivity Score

Hypotheses

Null Hypothesis (H₀) There is no difference between productivity scores before and after training.

Alternative Hypothesis (H₁) There is a significant difference between productivity scores before and after training.

Statistical Test

A Paired Samples t-Test was conducted to compare productivity scores before and after training.

Results

The analysis produced a t-statistic of t(29) = -65.80 and p-value p < 0.001.

Interpretation

The findings suggest that the training program was effective at changing employee productivity levels. Employees scored higher after training compared to before training.

library(readxl)
library(ggplot2)

# Dataset 1- Between Subjects

PROJECTDS1 <- read_excel("C:/Final Project/PROJECTDS1.xlsx")

# Creating Table and conducting the Chi Squared test.
burnout_table <- table(
  PROJECTDS1$WorkMode,
  PROJECTDS1$BurnoutLevel
)

burnout_table

##         
##          High Low Medium
##   Hybrid   10   0      5
##   Remote    3   9      3

chisq.test(burnout_table)

## Warning in chisq.test(burnout_table): Chi-squared approximation may be
## incorrect

## 
##  Pearson's Chi-squared test
## 
## data:  burnout_table
## X-squared = 13.269, df = 2, p-value = 0.001314

# Not Creating Histograms as all the tests are going to be conducted regardless of Histograms.

# Creating Scatterplot for Pearson's Correlation Test.
plot(
  PROJECTDS1$JobSatisfaction,
  PROJECTDS1$ProductivityScore,
  main = "Job Satisfaction vs Productivity",
  xlab = "Job Satisfaction",
  ylab = "Productivity Score",
  pch = 19,
  col = "blue"
)

# Conducting Pearson's Correlation Test.
cor.test(
  PROJECTDS1$JobSatisfaction,
  PROJECTDS1$ProductivityScore,
  method = "pearson"
)

## 
##  Pearson's product-moment correlation
## 
## data:  PROJECTDS1$JobSatisfaction and PROJECTDS1$ProductivityScore
## t = 35.226, df = 28, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  0.9765547 0.9947666
## sample estimates:
##       cor 
## 0.9889048

# Creating Boxplot for Independent Samples T-test.
boxplot(
  ProductivityScore ~ WorkMode,
  data = PROJECTDS1,
  main = "Productivity by Work Mode",
  xlab = "Work Mode",
  ylab = "Productivity Score",
  col = c("skyblue", "lightgreen")
)

# Conducting the Independent Samples T-test.
t.test(
  ProductivityScore ~ WorkMode,
  data = PROJECTDS1,
  var.equal = TRUE
)

## 
##  Two Sample t-test
## 
## data:  ProductivityScore by WorkMode
## t = -8.1362, df = 28, p-value = 7.393e-09
## alternative hypothesis: true difference in means between group Hybrid and group Remote is not equal to 0
## 95 percent confidence interval:
##  -20.94621 -12.52046
## sample estimates:
## mean in group Hybrid mean in group Remote 
##             71.33333             88.06667

# Dataset 2- Within Subjects

PROJECTDS2 <- read_excel("C:/Final Project/PROJECTDS2.xlsx")

# Creating Boxplot for Dependent Samples T-test.
boxplot(
  PROJECTDS2$Productivity_Before,
  PROJECTDS2$Productivity_After,
  names = c("Before", "After"),
  main = "Productivity Before and After Training",
  ylab = "Productivity Score",
  col = c("orange", "lightgreen")
)

# Conducting the Dependent Samples T-test.
t.test(
  PROJECTDS2$Productivity_Before,
  PROJECTDS2$Productivity_After,
  paired = TRUE
)

## 
##  Paired t-test
## 
## data:  PROJECTDS2$Productivity_Before and PROJECTDS2$Productivity_After
## t = -65.802, df = 29, p-value < 2.2e-16
## alternative hypothesis: true mean difference is not equal to 0
## 95 percent confidence interval:
##  -8.283021 -7.783646
## sample estimates:
## mean difference 
##       -8.033333