About Dataset¶

Compiled from the National Center of Education Statistics Annual Digest. Specifically, Table 330.20: Average undergraduate tuition and fees and room and board rates charged for full-time students in degree-granting postsecondary institutions, by control and level of institution and state or jurisdiction.

Dataset Link - Kaggle Link

Data Dictionary -¶

Year - The Digest year this information comes from
State - The U.S. State
Type - Type of University, Private or Public and in-state or out-of-state. Private colleges charge the same for in/out of state
Length - Whether the college mainly offers 2-year (Associates) or 4-year (Bachelors) programs
Expenses - The Expense being described, tuition/fees or on-campus living expenses
Value - The average cost for this particular expense, in USD ($)

Installing dependency¶

👉Ignore It if already installed

1. !pip install numpy
2. !pip install pandas
3. !pip install matplotlib
4. !pip install seaborn

step -1 Data Preprocessing and Cleaning¶

Importing Required library¶

In [1]:

# perform linear operations
import numpy as np

# Data manipulation
import pandas as pd

#Data Visualization
import matplotlib.pyplot as plt
import seaborn as sns

# Remove warnings
import warnings
warnings.filterwarnings('ignore')

In [2]:

#Load the dataset
us_data=pd.read_csv(r"C:\Users\Lenovo\Downloads\content\US undergrad data\nces330_20.csv")

# Print top 5 rows
us_data.head()

Out[2]:

	Year	State	Type	Length	Expense	Value
0	2013	Alabama	Private	4-year	Fees/Tuition	13983
1	2013	Alabama	Private	4-year	Room/Board	8503
2	2013	Alabama	Public In-State	2-year	Fees/Tuition	4048
3	2013	Alabama	Public In-State	4-year	Fees/Tuition	8073
4	2013	Alabama	Public In-State	4-year	Room/Board	8473

In [3]:

# check for shape
us_data.shape

Out[3]:

(3548, 6)

From above cell we see that the dataset contains 3548 observations and 6 columns

In [4]:

#Check info of each colummn
us_data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3548 entries, 0 to 3547
Data columns (total 6 columns):
 #   Column   Non-Null Count  Dtype 
---  ------   --------------  ----- 
 0   Year     3548 non-null   int64 
 1   State    3548 non-null   object
 2   Type     3548 non-null   object
 3   Length   3548 non-null   object
 4   Expense  3548 non-null   object
 5   Value    3548 non-null   int64 
dtypes: int64(2), object(4)
memory usage: 166.4+ KB

From above cell we see that there are 4 object column and 2 integer

In [5]:

# Checking null values
us_data.isnull().sum()

Out[5]:

Year       0
State      0
Type       0
Length     0
Expense    0
Value      0
dtype: int64

From above cell we see that there are no missing values in our dataset

In [6]:

# check for duplicate
us_data.duplicated().sum()

Out[6]:

From above cell we see that there are no duplicates present in our dataset

Step -2 Data Analysis¶

Let's Check the Distribution of each column¶

In [7]:

us_data

Out[7]:

	Year	State	Type	Length	Expense	Value
0	2013	Alabama	Private	4-year	Fees/Tuition	13983
1	2013	Alabama	Private	4-year	Room/Board	8503
2	2013	Alabama	Public In-State	2-year	Fees/Tuition	4048
3	2013	Alabama	Public In-State	4-year	Fees/Tuition	8073
4	2013	Alabama	Public In-State	4-year	Room/Board	8473
...	...	...	...	...	...	...
3543	2021	Wyoming	Public In-State	2-year	Fees/Tuition	3987
3544	2021	Wyoming	Public In-State	4-year	Room/Board	9799
3545	2021	Wyoming	Public Out-of-State	2-year	Fees/Tuition	9820
3546	2021	Wyoming	Public Out-of-State	4-year	Fees/Tuition	14710
3547	2021	Wyoming	Public Out-of-State	4-year	Room/Board	9799

3548 rows × 6 columns

`Year`¶

In [14]:

ax=sns.countplot(x='Year',data=us_data)
for label in ax.containers:
    ax.bar_label(label)
plt.title("Various years available in the dataset")
plt.show()

No description has been provided for this image

From above bar plot we see that the year range is start from 2013 and end upto 2021

`State`¶

In [19]:

plt.figure(figsize=(10,5))
ax=sns.countplot(x='State',data=us_data)
for label in ax.containers:
    ax.bar_label(label)
plt.title("Various states available in the dataset")
plt.xticks(rotation=90)
plt.show()

Above plot indicates that there are various states in our data or all the states of US

`Type`¶

In [21]:

plt.figure(figsize=(10,5))
ax=sns.countplot(x='Type',data=us_data)
for label in ax.containers:
    ax.bar_label(label)
plt.title("Various Types of Universities in our data")
plt.xticks(rotation=45)
plt.show()

The distribution of universities across the United States reveals three distinct categories: "Private," "Public In-State," and "Public Out-of-State." Notably, the dataset indicates the prevalence of 905 "private" universities, 1296 "public in-state" institutions, and 1347 "public out-of-state" establishments throughout the country. This data underscores the diversity in the higher education landscape, with various institutions offering educational opportunities to students nationwide. The presence of a larger number of public universities, both in-state and out-of-state, suggests the significance of accessible and affordable education within the United States.

`Length`¶

In [22]:

plt.figure(figsize=(10,5))
ax=sns.countplot(x='Length',data=us_data)
for label in ax.containers:
    ax.bar_label(label)
plt.title("Types of program available")
plt.xticks(rotation=45)
plt.show()

The analysis of university programs in the United States highlights the prevalence of two primary categories: "4-year" and "2-year" programs. The dataset demonstrates a substantial presence of 2672 records affiliated with "4-year" programs and 876 observations associated with "2-year" programs.

`Expense`¶

In [24]:

plt.figure(figsize=(10,5))
ax=sns.countplot(x='Expense',data=us_data)
for label in ax.containers:
    ax.bar_label(label)
plt.title("Types of Expense")
plt.xticks(rotation=45)
plt.show()

The bar plot elucidates the distinct types of expenses incurred by students in the United States, namely "Fees/Tuition" and "Room/Board." Notably, the data reveals that there are 2198 records associated with "Fees/Tuition" expenses, while 1350 records correspond to "Room/Board" expenditures.

`Value`¶

In [29]:

sns.histplot(us_data['Value'],kde=True)
plt.title("distribution of cost")
plt.show()

Let's ask some question from the data¶

How do the expenses differ between public and private universities in various states?¶

In [30]:

us_data

Out[30]:

	Year	State	Type	Length	Expense	Value
0	2013	Alabama	Private	4-year	Fees/Tuition	13983
1	2013	Alabama	Private	4-year	Room/Board	8503
2	2013	Alabama	Public In-State	2-year	Fees/Tuition	4048
3	2013	Alabama	Public In-State	4-year	Fees/Tuition	8073
4	2013	Alabama	Public In-State	4-year	Room/Board	8473
...	...	...	...	...	...	...
3543	2021	Wyoming	Public In-State	2-year	Fees/Tuition	3987
3544	2021	Wyoming	Public In-State	4-year	Room/Board	9799
3545	2021	Wyoming	Public Out-of-State	2-year	Fees/Tuition	9820
3546	2021	Wyoming	Public Out-of-State	4-year	Fees/Tuition	14710
3547	2021	Wyoming	Public Out-of-State	4-year	Room/Board	9799

3548 rows × 6 columns

In [37]:

plt.figure(figsize=(14, 5))
sns.barplot(x='State', y='Value', hue='Type', data=us_data[us_data.Expense=='Fees/Tuition'], ci=None)
plt.title('Average Expenses of Tuition fees for Different Types of Universities in Each State')
plt.xticks(rotation=45)
plt.xlabel('State')
plt.ylabel('Average Expense Value')
plt.show()

Exploratory Data Analysis on US Undergrad Data in Python