2.
1 Introduction to Data Analysis
Data analysis is a crucial phase of the research process as it transforms
raw data into meaningful insights that support decision-making. This
chapter presents a detailed analysis of the collected college placement
data using appropriate statistical and analytical tools. The analysis
primarily focuses on understanding the influence of academic
performance (CGPA), skills, and internships on placement outcomes.
Various pivot tables and charts are used to identify patterns, trends, and
relationships among the variables.
2.2 Descriptive Statistics
Measures of Central Tendency
Measures of central tendency were used to identify the central value
around which the CGPA data is distributed.
Measur Variabl Valu
es es e
Mean CGPA 7.532
Median CGPA 7.550
Mode CGPA 9.410
Interpretation: The analysis of central tendency indicates that the mean
CGPA is 7.53, representing the average performance across the sample.
The mean and median are closely aligned, suggesting a relatively
balanced distribution of academic scores.
Measures of Dispersion
Measures of dispersion describe the extent to which data values are
spread or scattered around the central value.
Variabl Valu
Measure
e e
Minimum CGPA 4.54
Maximum CGPA 10.46
Variabl Valu
Measure
e e
Range CGPA 5.92
Standard
CGPA 1.47
Deviation
Interpretation: The minimum CGPA recorded is 4.54, while the
maximum is 10.46, indicating a significant difference in academic levels.
The standard deviation of 1.47 shows that most students’ CGPAs are
moderately spread around the average.
Distribution Characteristics
Distribution characteristics indicate how the data values are arranged
around the central value.
Variabl Valu
Measure
e e
Skewne -
CGPA
ss 0.033
-
Kurtosis CGPA
1.110
Interpretation: The skewness of -0.033 indicates a nearly symmetric
distribution with a slight lean towards higher CGPA values. The negative
kurtosis (-1.11) suggests a platykurtic distribution, meaning the data has
thinner tails and is flatter than a normal distribution.
2.3 Outlier Analysis
Outlier analysis is the process of identifying extreme values in a dataset
that differ significantly from the majority of observations.
Value -
Measure
CGPA
Q1 (First Quartile) 6.29
Value -
Measure
CGPA
Q3 (Third Quartile) 8.77
Interquartile Range
2.48
(IQR)
Lower Bound (LB) 2.57
Upper Bound (UB) 12.49
Minimum Value 4.54
Maximum Value 10.46
Interpretation: Half of the students have a CGPA between 6.29 and 8.77.
Since the minimum (4.54) and maximum (10.46) fall within the Lower and
Upper bounds, there are no extreme outliers in the CGPA column for this
dataset.
2.4 Univariate, Bivariate, Multivariate Analysis
Univariate Analysis
Univariate analysis focuses on examining the distribution and
characteristics of a single variable.
Value -
Measure
CGPA
X (Selected
7.532
Value)
Mean 7.532
Standard
1.470
Deviation
Value -
Measure
CGPA
P (CGPA < X) 0.50
P (CGPA > X) 0.50
Interpretation: Approximately 50% of the students have a CGPA below
the mean. Unlike the Capstone's potential savings analysis, the academic
data is evenly spread around the average.
2.5 Tabular Analysis and Data Visualisation
[1] What is the percentage distribution of respondents across
placement outcomes?
Row Count of Percenta
Labels Students ge
Not Placed 8,341 83.41%
Placed 1,659 16.59%
Grand
10,000 100%
Total
Interpretation: The majority of respondents (83%) were not placed,
indicating a highly competitive environment or high selection standards
within the dataset.
[2] How does Average CGPA vary by Internship Experience?
Internship Average of
Experience CGPA
No 7.547
Yes 7.510
Interpretation: Interestingly, the average CGPA is virtually identical for
students with and without internship experience, suggesting that
internships are not necessarily linked to higher academic scores in this
group.
[3] How does Academic Performance influence Placement
success?
Academic Not Place
Performance Placed d
High (8-10) 2,550 506
Medium (4-7) 3,356 641
Low (1-3) 2,437 512
Interpretation: Similar to the life-cycle approach in the capstone study ,
placement success is distributed across performance levels, but students
with higher academic performance ratings represent a significant portion
of the successful pool.