STAT 124 - Exploratory Data Anaysis

Introduction to Statistics

  • Course: STAT 124

  • Topic: Exploratory Data Analysis (EDA)

  • Institution: UBC

EDA Overview

Command Example

  • A data set was collected through a survey asking students to express their level of agreement with the statement: "I am afraid that I will not be able to get a satisfactory job after graduation."

  • The responses were recorded on a 5-point Likert scale: 1 (Strongly Disagree) to 5 (Strongly Agree).

  • Data representation in R:

    • Survey2 <- structure(c(1L, 5L, 1L, 4L, 1L, 1L, 1L, 4L, 2L, 1L, 1L, 1L, 4L, 1L, 2L, 3L, 5L, 1L, 1L, 3L, 1L, 3L, 1L, 2L, 3L, 1L, 5L, 2L, 2L, 3L, 3L, 5L, 3L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 2L, 1L, 1L, 1L, 4L, 2L, 2L, 2L, 1L, 3L, 4L, 2L, 2L, 1L, 3L, 1L, 4L, 2L, 1L, 2L, 3L, 5L, 2L, 2L, 1L, 3L, 2L, 1L, 3L, 4L, 1L, 1L, 2L, 1L, 2L, 1L, 1L, 5L, 5L, 4L, 1L, 2L, 1L, 1L, 5L, 2L, 4L,3L, 1L), levels = c("Agree", "Disagree", "Neutral", "Str. Agree", "Str. Disagree"), class = "factor")

  • The resulting graphical display of responses is contained in the files:

    • copilot.pdf

    • gemini.pdf

Report 1: Visualization of Likert-Scale Responses

  • Platform: Microsoft Copilot

  • Date: January 10, 2026

Introduction

  • The survey aimed to measure students' anxiety regarding job prospects after graduation based on their agreement with the statement: “I am afraid that I will not be able to get a satisfactory job after graduation.”

  • Responses were categorized using a 5-point Likert scale:

    • Categories:

    • Strongly Disagree

    • Disagree

    • Neutral

    • Agree

    • Strongly Agree

Visualization

  • Utilized a bar chart to visualize the frequency of each Likert category.

  • Notes: The bar chart enables a straightforward comparison of student agreement levels.

Report 1 - Page 2: Visualization Results

Figure 1

  • Title: Distribution of student responses regarding fear of not obtaining a satisfactory job after graduation.

Report 2: Analysis of Student Job Prospect Anxiety

  • Platform: Google Gemini

  • Date: January 10, 2026

Introduction

  • The aim was to analyze student responses to the aforementioned statement using a 5-point Likert scale.

Data Summary

  • Total Responses: 95

  • Frequency distribution of responses is illustrated in Table 1:

    • Table 1: Frequency Distribution of Likert Responses
      | Response Level | Frequency |
      |----------------------|-----------|
      | Strongly Agree | 9 |
      | Agree | 39 |
      | Neutral | 13 |
      | Disagree | 26 |
      | Strongly Disagree | 8 |
      | Total | 95 |

Visual Representation

  • Figure 1: Illustrates the distribution of responses showing that the mode of the dataset is “Agree,” indicating significant anxiety among students regarding job prospect insecurity.

Key Findings

  • High Anxiety: Approximately 50.5% (48 out of 95) of students expressed agreement with the statement.

  • Low Confidence: Only 35.8% (34 out of 95) disagreed or strongly disagreed.

  • Neutrality: 13.7% of respondents remained neutral, suggesting uncertain or ambiguous feelings towards employment prospects.

Report 2 - Page 2: Visual Analysis of Responses

Figure 1

  • Title: Distribution of Student Agreement Levels

  • Frequency distribution displayed with categories:

    • Levels: Str. Agree, Agree, Neutral, Disagree, Str. Disagree

Additional Graphs

Bar Chart Options

  • Bar Chart 1: Default display, decreasing order of frequency.

    • Categories: Agree, Disagree, Neutral, Str. Agree, Str. Disagree

  • Bar Chart 2: Ordered according to the Likert Scale:

    • Categories: Str. Agree, Agree, Neutral, Disagree, Str. Disagree

Pie Chart (For Amusement Only)

  • Caution: The ordering of the Likert scale is not respected, has various interpretive limitations.

Nonresponses Representation

  • Visualization of responses including nonresponses.

EDA - Graphics Example 2

Command Example

  • Discussion regarding octane ratings of gasoline samples.

  • Raw data listed: 88.5, 95.6, 88.3, ...

  • Resulting reports found in:

    • copilot_gas.pdf

    • gemini_gas.pdf

Report 1: Graphical Analysis of Octane Rating Data

  • Platform: Microsoft Copilot

Introduction

  • Provides graphical summary of a dataset with octane ratings from gasoline samples.

  • Note: One extreme outlier (990.4) identified, distorting histogram.

Visualization

  • Zoomed-In Histogram: Excludes outlier for clearer distribution view.

  • Majority of values found between approximately 83 and 100, with concentration in high 80s and low 90s.

Report 1 - Page 2 Notes

  • Retained outlier (990.4) in dataset but excluded from visualization.

  • A full-range histogram can be provided if needed.

Report 2: Statistical Analysis of Gasoline Octane Ratings

  • Platform: Google Gemini

  • Date: January 10, 2026

Introduction

  • Overview of octane ratings from 79 gasoline samples.

  • Outlier identified and corrected from 990.4 to 90.4.

Numerical Summary

  • Summary statistics presented in Table 1:

    • Table 1: Summary Statistics for Octane Ratings
      | Statistic | Value |
      |-----------------------|-----------|
      | Sample Size (n) | 79 |
      | Mean (\bar{x}) | 90.56 |
      | Median (\tilde{x}) | 90.40 |
      | Standard Deviation (s)| 2.65 |
      | Minimum | 83.40 |
      | Maximum | 100.30 |

Graphical Representation

  • Utilized two primary visualizations: Histogram and Box Plot.

3.1 Frequency Distribution (Histogram)
  • A graph illustrating the distribution shape, showing unimodal and approximately normal patterns with high frequencies between 90.0 and 92.0.

3.2 Variability and Outliers (Box Plot)
  • Displayed five-number summary including Interquartile Range (IQR) and highlighting extreme values.

Report 2 - Page 2

Figure 1

  • Visual representation of frequency and data spread.

Conclusion

  • The analysis indicates a consistent octane quality in gasoline samples, predominantly between 88.0 and 93.0, with a few outliers potentially needing further scrutiny.