Lecture 1 – Data 100, Fall 2023¶

Data 100, Fall 2023

Acknowledgments Page

Software Packages¶

We will be using a wide range of different Python software packages. To install and manage these packages we will be using the Conda environment manager. The following is a list of packages we will routinely use in lectures and homeworks:

In [1]:
# Linear algebra, probability
import numpy as np

# Data manipulation
import pandas as pd

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

# Interactive visualization library
import plotly.offline as py
py.init_notebook_mode(connected=True)
import plotly.graph_objs as go
import plotly.figure_factory as ff
import plotly.express as px

We will learn how to use all of the technologies used in this demo.

For now, just sit back and think critically about the data and our guided analysis.

1. Starting with a Question: Who are you (the students of Data 100)?¶

No description has been provided for this image

This is a pretty vague question but let's start with the goal of learning something about the students in the class.

Here are some "simple" questions:

  1. How many students do we have?
  2. What are your majors?
  3. What year are you?
  4. How did your major enrollment trend change over time?

2. Data Acquisition and Cleaning¶

In Data 100 we will study various methods to collect data.

No description has been provided for this image

To answer this question, I downloaded the course roster and extracted everyone's names and majors.

In [2]:
# pd stands for pandas, which we will learn starting from next lecture
# Some pandas syntax shared with data8's datascience package
majors = pd.read_csv("data/majors.csv")
names = pd.read_csv("data/names.csv")

3. Exploratory Data Analysis¶

In Data 100 we will study exploratory data analysis and practice analyzing new datasets.

No description has been provided for this image

I didn't tell you the details of the data! Let's check out the data and infer its structure. Then we can start answering the simple questions we posed.

Peeking at the Data¶

In [3]:
# Let's peek at the first 20 rows of the majors dataframe

majors.head(20)
Out[3]:
Majors Terms in Attendance
0 Letters & Sci Undeclared UG 3
1 Data Science BA 7
2 Economics BA 8
3 Letters & Sci Undeclared UG 5
4 Economics BA 7
5 Computer Science BA, Economics BA 7
6 Business Administration BS, Letters & Sci Unde... 5
7 Letters & Sci Undeclared UG 5
8 Microbial Biology BS 5
9 Letters & Sci Undeclared UG 8
10 Letters & Sci Undeclared UG 3
11 Letters & Sci Undeclared UG 5
12 Data Science BA, Society and Environment BS 5
13 Letters & Sci Undeclared UG 3
14 Letters & Sci Undeclared UG, Microbial Biology BS 5
15 Letters & Sci Undeclared UG, Nutritional Scien... 5
16 Letters & Sci Undeclared UG 3
17 Chemistry BS 5
18 Applied Mathematics BA 7
19 Letters & Sci Undeclared UG 3
In [4]:
# Let's peek at the first 5 rows (default) of the names dataframe

names.head()
Out[4]:
Name Role
0 Emily Student
1 Zoe Student
2 Michelle Student
3 JAMES Student
4 Jenny Student

What is one potential issue we may need to address in this data?¶

Answer: Some names appear capitalized.

In the above sample we notice that some of the names are capitalized and some are not. This will be an issue in our later analysis so let's convert all names to lower case.

In [5]:
names['Name'] = names['Name'].str.lower()
In [6]:
names.head()
Out[6]:
Name Role
0 emily Student
1 zoe Student
2 michelle Student
3 james Student
4 jenny Student

Exploratory Data Analysis on names dataset¶

What is the most common first letter in names? What is its distribution?¶

In [7]:
# Below are the most common, in descending frequency

first_letter = names['Name'].str[0].value_counts()
first_letter.head()
Out[7]:
Name
a    192
j    137
s    130
m     97
e     72
Name: count, dtype: int64
In [8]:
# Let's visualize this first letter distribution 

plt.bar(first_letter.index, first_letter.values)
plt.xlabel('First Letter')
plt.ylabel('Frequency')
plt.title('First Letter Frequency Distribution')
plt.show()
No description has been provided for this image

In the United States, "J" and "A" names are the most popular first initials. Seems like our visualization also reflects this!

What is the distribution of the length of names?¶

In [9]:
name_lengths = names['Name'].str.len()

plt.hist(name_lengths, bins=range(min(name_lengths), max(name_lengths) + 2), edgecolor='black')
plt.xlabel('Name Length')
plt.ylabel('Frequency')
plt.title('Distribution of Length of Names')
average_length = name_lengths.sum() / len(name_lengths)
plt.axvline(average_length, color='red', linestyle='dashed', linewidth=1, label=f'Average: {average_length:.2f}')
plt.legend()
plt.xticks(range(min(name_lengths), max(name_lengths) + 1))
plt.show()
No description has been provided for this image

The average length of names in the United States is also around 6 letters!

How many records do we have?¶

In [10]:
print(len(names))
print(len(majors))

1291
1291

Based on what we know of our class, each record is most likely a student.

Understanding the structure of data¶

It is important that we understand the meaning of each field and how the data is organized.

In [11]:
names.head()
Out[11]:
Name Role
0 emily Student
1 zoe Student
2 michelle Student
3 james Student
4 jenny Student

What is the meaning of the Role field?¶

Answer: Understanding the meaning of an attribute can be achieved by looking at the types of data it contains (in particular, the counts of its unique values).

We use the value_counts() function in pandas:

In [12]:
names['Role'].value_counts().to_frame()  # counts of unique Roles
Out[12]:
count
Role
Student 1290
#REF! 1

It appears that one student has an erroneous role given as "#REF!". What else can we learn about this student? Let's see their name.

In [13]:
# Boolean index to find rows where Role is #REF!
names[names['Role'] == "#REF!"]
Out[13]:
Name Role
737 #ref! #REF!

Though this single bad record won't have much of an impact on our analysis, we can clean our data by removing this record.

In [14]:
names = names[names['Role'] != "#REF!"]

Double check: Let's double check that our record removal only removed the single bad record.

In [15]:
names['Role'].value_counts().to_frame()  # Again, counts of unique Roles
Out[15]:
count
Role
Student 1290

Remember we loaded in two files. Let's explore the fields of majors and check for bad records:

Exploratory Data Analysis on majors dataset¶

In [16]:
majors.columns   # Get column names
Out[16]:
Index(['Majors', 'Terms in Attendance'], dtype='object')
In [17]:
majors['Terms in Attendance'].value_counts().to_frame()
Out[17]:
count
Terms in Attendance
5 539
7 351
3 171
G 124
8 75
6 22
4 8
#REF! 1

It looks like numbers represent semesters, G represents graduate students. But we do still have a bad record:

In [18]:
majors[majors['Terms in Attendance'] == "#REF!"]
Out[18]:
Majors Terms in Attendance
597 #REF! #REF!
In [19]:
majors = majors[majors['Terms in Attendance'] != "#REF!"]
majors['Terms in Attendance'].value_counts().to_frame()
Out[19]:
count
Terms in Attendance
5 539
7 351
3 171
G 124
8 75
6 22
4 8

Detail: The deleted majors record number is different from the record number of the bad names record. So while the number of records in each table matches, the row indices don't match, so we'll have to keep these tables separate in order to do our analysis.

Summarizing the Data¶

We will often want to numerically or visually summarize the data. The describe() method provides a brief high level description of our data frame.

In [20]:
names.describe()
Out[20]:
Name Role
count 1290 1290
unique 933 1
top ethan Student
freq 11 1290

Q: What do you think top and freq represent?

Answer: top: most frequent entry, freq: the frequency of that entry

In [21]:
majors.describe()
Out[21]:
Majors Terms in Attendance
count 1290 1290
unique 183 7
top Letters & Sci Undeclared UG 5
freq 353 539

4. Analysis: Understanding the World¶

No description has been provided for this image

What are your majors?¶

What are the top majors:

In [22]:
majors_count = (       # Method chaining in pandas
    majors['Majors']
    .value_counts()
    .sort_values(ascending=False) # Highest first
    .to_frame()
    .head(20)          # Get the top 20
)

majors_count
Out[22]:
count
Majors
Letters & Sci Undeclared UG 353
Data Science BA 129
Computer Science BA 124
Electrical Eng & Comp Sci BS 69
Economics BA 65
Cognitive Science BA 29
Electrical Eng & Comp Sci MEng 25
Applied Mathematics BA 24
Civil Engineering BS 22
Mol Sci & Software Engin MMSSE 19
Business Administration BS, Letters & Sci Undeclared UG 19
Data Science BA, Economics BA 17
Chemical Engineering BS 15
Development Engineering MDE 13
Public Policy MPP 12
Computer Science BA, Data Science BA 11
Info Mgmt & Systems MIMS 10
Molecular & Cell Biology BA 10
Bioengineering BS 9
Public Health BA 8

We will often use visualizations to make sense of data¶

In Data 100, we will deal with many different kinds of data (not just numbers) and we will study techniques to describe types of data.

How can we summarize the Majors field? A good starting point might be to use a bar plot:

In [23]:
# Interactive using plotly

fig = px.bar(majors_count.loc[::-1], orientation='h')
fig.update_layout(showlegend=False, 
                  xaxis_title='Count', 
                  yaxis_title='Major',
                  autosize=False, 
                  width=800, 
                  height=500)

What year are you?¶

In [24]:
fig = px.histogram(majors['Terms in Attendance'].sort_values(),
                   histnorm='probability')

fig.update_layout(showlegend=False,
                  xaxis_title="Term",
                  yaxis_title="Fraction of Class",
                  autosize=False, 
                  width=800, 
                  height=250)
In [25]:
# Replacing terms in attendance data with the degree objective
majors.loc[majors.loc[:, 'Terms in Attendance'] != 'G', 'Terms in Attendance'] = 'Undergraduate'
majors.loc[majors.loc[:, 'Terms in Attendance'] == 'G', 'Terms in Attendance'] = 'Graduate'


majors.rename(columns={'Terms in Attendance': 'Ungrad Grad'}, inplace=True)

majors.describe()
Out[25]:
Majors Ungrad Grad
count 1290 1290
unique 183 2
top Letters & Sci Undeclared UG Undergraduate
freq 353 1166

1. New Questions¶

  • What is the ratio between graduate and undergraduate students in Data 100, and how does it compare with campus distribution?

  • What is the proportion of different majors in Data 100, and how does it compare with historical campus trends?

No description has been provided for this image

We often ask this question because we want to improve the data science program here in Berkeley, especially since it has now grown into a new college—College of Computing, Data Science, and Society—Berkeley's first new college in 50 years.

How could we answer this question?¶

In [26]:
print(majors.columns)
print(names.columns)

Index(['Majors', 'Ungrad Grad'], dtype='object')
Index(['Name', 'Role'], dtype='object')

We don't have the data.¶

Where can we get the data?

No description has been provided for this image

UC Berkeley Student Headcount by Major and Demographics¶

Data comes from UC Berkeley's data warehouse reporting system, Cal Answers. Headcount figures are based on the Fall and Spring census snapshots, which occur at the end of the 5th week of the term.

2. Acquire data programmatically¶

Note 1: In the following, we download the data programmatically to ensure that the process is reproducible.

Note 2: We also load the data directly into Python.

In Data 100 we will think a bit more about how we can be efficient in our data analysis to support processing large datasets.

In [27]:
url = "https://docs.google.com/spreadsheets/d/1J7tz3GQLs3M6hFseJCE9KhjVhe4vKga8Q2ezu0oG5sQ/gviz/tq?tqx=out:csv"

university_majors = pd.read_csv(url, 
                                usecols = ['Academic Yr', 'Semester', 'Ungrad Grad', 
                                           'Entry Status', 'Major Short Nm', 'Student Headcount'])

3. Exploratory Data Analysis on Campus Data¶

No description has been provided for this image

In [28]:
# Examining the data
university_majors
Out[28]:
Academic Yr Semester Ungrad Grad Entry Status Major Short Nm Student Headcount
0 2013-14 Fall Graduate Graduate Education 327
1 2013-14 Fall Graduate Graduate Special Education 14
2 2013-14 Fall Graduate Graduate Science & Math Education 16
3 2013-14 Fall Graduate Graduate Chemical Engineering 132
4 2013-14 Fall Graduate Graduate Chemistry 404
... ... ... ... ... ... ...
7278 2022-23 Spring Undergraduate Transfer Entrant Nut Sci-Physio & Metabol 20
7279 2022-23 Spring Undergraduate Transfer Entrant Nutritional Sci-Dietetics 3
7280 2022-23 Spring Undergraduate Transfer Entrant Nutritional Sci-Toxicology 3
7281 2022-23 Spring Undergraduate Transfer Entrant Genetics & Plant Biology 10
7282 2022-23 Spring Undergraduate Transfer Entrant Microbial Biology 49

7283 rows × 6 columns

The data is reported on a semester basis. We will aggregate data across different semesters in a year by taking average of Fall and Spring semester enrollment information.

In [29]:
# Reporting student data based on academic year
university_majors = (university_majors.groupby(
    ['Academic Yr', 'Ungrad Grad', 'Entry Status', 'Major Short Nm'], as_index = False)[["Student Headcount"]]
                     .mean()
                    )

university_majors
Out[29]:
Academic Yr Ungrad Grad Entry Status Major Short Nm Student Headcount
0 2013-14 Graduate Graduate African American Studies 30.0
1 2013-14 Graduate Graduate Ag & Resource Economics 71.5
2 2013-14 Graduate Graduate Anc Hist & Medit Archae 15.0
3 2013-14 Graduate Graduate Anthropology 88.5
4 2013-14 Graduate Graduate Applied Mathematics 16.5
... ... ... ... ... ...
3735 2022-23 Undergraduate Transfer Entrant Spanish and Portuguese 12.0
3736 2022-23 Undergraduate Transfer Entrant Statistics 35.0
3737 2022-23 Undergraduate Transfer Entrant Sustainable Environ Dsgn 4.5
3738 2022-23 Undergraduate Transfer Entrant Theater & Perf Studies 44.0
3739 2022-23 Undergraduate Transfer Entrant Urban Studies 16.0

3740 rows × 5 columns

What is the historical distribution of graduate and undergraduate students at Berkeley?¶

In [30]:
university_grad_vs_ungrd = (university_majors.groupby(
    ['Academic Yr', 'Ungrad Grad'], as_index = False)[["Student Headcount"]]
                            .sum()
                           )

proportions = university_grad_vs_ungrd.pivot(index='Academic Yr', columns='Ungrad Grad', values='Student Headcount')
proportions['Total'] = proportions['Undergraduate'] + proportions['Graduate']
proportions['Undergrad Proportion'] = proportions['Undergraduate'] / proportions['Total']
proportions['Grad Proportion'] = proportions['Graduate'] / proportions['Total']


fig = px.bar(proportions.reset_index(), 
             x='Academic Yr', 
             y=['Undergraduate', 'Graduate'],
             title='Number of Grad vs. Undergrad Students',
             labels={'value': 'Number of Students'},
             color_discrete_map={'Undergraduate': 'blue', 'Graduate': 'orange'})

fig.update_layout(barmode='relative', autosize=False, width=800, height=600)
fig.show()

4.1. Ratio between graduate and undergraduate students in Data 100, and its comparison with campus distribution¶

In [31]:
data100_grad = majors['Ungrad Grad'].loc[majors['Ungrad Grad'] == 'Graduate'].count()

data100_undergrad = majors['Ungrad Grad'].loc[majors['Ungrad Grad'] == 'Undergraduate'].count()

print("Number of graduate students in Data 100: ", data100_grad)
print("Number of undergraduate students in Data 100: ", data100_undergrad)

Number of graduate students in Data 100:  124
Number of undergraduate students in Data 100:  1166
In [32]:
data100_row = {'Graduate':[data100_grad], 
               'Undergraduate':[data100_undergrad], 
               'Total':[data100_grad + data100_undergrad], 
               'Undergrad Proportion':[data100_undergrad / (data100_grad + data100_undergrad)],
               'Grad Proportion':[data100_grad / (data100_grad + data100_undergrad)], 
               }

new_row_df = pd.DataFrame(data100_row)

proportions.loc['Data 100'] = new_row_df.iloc[0]


fig = px.bar(proportions.reset_index(), 
             x='Academic Yr', 
             y=['Undergrad Proportion', 'Grad Proportion'],
             title='Proportions of Grad vs. Undergrad Students',
             labels={'value': 'Proportion'},
             color_discrete_map={'Undergrad Proportion': 'blue', 'Grad Proportion': 'orange'})


fig.update_layout(barmode='relative', autosize=False, width=800, height=600)
fig.show()

4.2. Proportion of different majors in Data 100, and their historical emrollment trends¶

In [33]:
data100_top_20_majors = (       # Method chaining in pandas
    majors['Majors']
    .value_counts()
    .sort_values(ascending=False) # Highest first
    .to_frame()
    .head(20)          # Get the top 20
)

major_trends = university_majors.groupby(['Academic Yr', 'Major Short Nm'], 
                                         as_index = False)[["Student Headcount"]].sum()


print("Top 20 majors at Berkeley in 2022-23")
major_trends[major_trends.loc[:, 'Academic Yr'] == '2022-23'].sort_values('Student Headcount', ascending=False).head(20)

Top 20 majors at Berkeley in 2022-23
Out[33]:
Academic Yr Major Short Nm Student Headcount
1993 2022-23 Letters & Sci Undeclared 10651.0
1983 2022-23 L&S Computer Science 2102.5
1932 2022-23 Electrical Eng & Comp Sci 2093.0
1894 2022-23 Business Administration 1645.5
1928 2022-23 Economics 1579.5
1984 2022-23 L&S Data Science 1325.5
2020 2022-23 Molecular & Cell Biology 1225.5
2011 2022-23 Mechanical Engineering 1208.0
1992 2022-23 Law (JD) 1023.0
1973 2022-23 Info & Data Science-MIDS 1021.5
2042 2022-23 Political Science 1005.0
1948 2022-23 Evening & Weekend MBA 919.0
2043 2022-23 Psychology 760.0
1901 2022-23 Chemistry 691.0
2057 2022-23 Sociology 663.0
1881 2022-23 Architecture 604.5
1900 2022-23 Chemical Engineering 595.0
1889 2022-23 Bioengineering 576.0
1912 2022-23 Cognitive Science 505.5
1879 2022-23 Applied Mathematics 497.0
In [34]:
print("Top 20 majors at Berkeley since 2013")
major_trends.groupby(['Major Short Nm'], as_index = False)[['Student Headcount']].sum().sort_values('Student Headcount', ascending=False).head(20)
     

Top 20 majors at Berkeley since 2013
Out[34]:
Major Short Nm Student Headcount
150 Letters & Sci Undeclared 101418.0
77 Electrical Eng & Comp Sci 18431.0
137 L&S Computer Science 14818.0
33 Business Administration 14302.5
72 Economics 14000.0
216 Political Science 10334.0
176 Mechanical Engineering 10193.5
149 Law (JD) 9645.5
95 Evening & Weekend MBA 7932.5
233 Sociology 6719.5
217 Psychology 6533.0
40 Chemical Engineering 6126.5
41 Chemistry 5941.5
13 Architecture 5680.5
85 English 5520.0
187 Molecular & Cell Biology 5431.0
125 Info & Data Science-MIDS 5105.0
26 Bioengineering 5061.0
128 Integrative Biology 5060.5
211 Physics 5055.0
In [35]:
print("Top 20 majors at Berkeley in Data 100")
print(data100_top_20_majors)

Top 20 majors at Berkeley in Data 100
                                                    count
Majors                                                   
Letters & Sci Undeclared UG                           353
Data Science BA                                       129
Computer Science BA                                   124
Electrical Eng & Comp Sci BS                           69
Economics BA                                           65
Cognitive Science BA                                   29
Electrical Eng & Comp Sci MEng                         25
Applied Mathematics BA                                 24
Civil Engineering BS                                   22
Mol Sci & Software Engin MMSSE                         19
Business Administration BS, Letters & Sci Undec...     19
Data Science BA, Economics BA                          17
Chemical Engineering BS                                15
Development Engineering MDE                            13
Public Policy MPP                                      12
Computer Science BA, Data Science BA                   11
Info Mgmt & Systems MIMS                               10
Molecular & Cell Biology BA                            10
Bioengineering BS                                       9
Public Health BA                                        8
In [36]:
data100_top_20_majors.index = data100_top_20_majors.index.str.rsplit(' ', n=1).str[0]
print("Top 20 majors at Berkeley in Data 100")
print(data100_top_20_majors)

Top 20 majors at Berkeley in Data 100
                                                    count
Majors                                                   
Letters & Sci Undeclared                              353
Data Science                                          129
Computer Science                                      124
Electrical Eng & Comp Sci                              69
Economics                                              65
Cognitive Science                                      29
Electrical Eng & Comp Sci                              25
Applied Mathematics                                    24
Civil Engineering                                      22
Mol Sci & Software Engin                               19
Business Administration BS, Letters & Sci Undec...     19
Data Science BA, Economics                             17
Chemical Engineering                                   15
Development Engineering                                13
Public Policy                                          12
Computer Science BA, Data Science                      11
Info Mgmt & Systems                                    10
Molecular & Cell Biology                               10
Bioengineering                                          9
Public Health                                           8
In [37]:
fig = px.line(major_trends[major_trends["Major Short Nm"].isin(data100_top_20_majors.index)], 
                        x = "Academic Yr", y = "Student Headcount", color = "Major Short Nm")

fig.update_layout(autosize=False, width=800, height=600)
fig.show()
In [38]:
data100_top_19_majors = data100_top_20_majors.iloc[1:,:]

fig = px.line(major_trends[major_trends["Major Short Nm"].isin(data100_top_19_majors.index)], 
                        x = "Academic Yr", y = "Student Headcount", color = "Major Short Nm")

fig.update_layout(autosize=False, width=800, height=600)
fig.show()