Welcome to National-University-Ranking-App Discussions!

[TelRich]

👋 Welcome!

We’re using Discussions as a place to connect with other members of our community. We hope that you:

• Ask questions you’re wondering about.
• Share ideas.
• Engage with other community members.
• Welcome others and are open-minded. Remember that this is a community we
  build together 💪.

To get started, comment below with an introduction of yourself and tell us about what you do with this community.

[olufemilite]

@TelRich Here's the code for my data cleaning task;
#Changing column headers to lower case and changing spaces to underscore
df_nur.rename(columns={"Name":"name","Location":"location","Rank":"rank","Description":"description","Tuition and fees":"tuition_and_fees","In-state":"in_state","Undergrad enrollment":"undergrad_enrollment"}, inplace=True)
df_nur.head()

[TelRich]

Seen. Will implement. Thank you

[Mueeshassan]

States with the highest number of schools

states_nur = nur_data['state_full'].value_counts(ascending=True)
print(states_nur)
print(len(states_nur))
states_nur[states_nur > 10].sum() / states_nur.sum()

ax = states_nur.plot(kind='barh', figsize=(20, 15))
ax.set_xlabel('Number of Schools', fontsize= 30)
ax.set_ylabel('States', fontsize= 30)

Five states (California, Newyork, Massachusetts, Pennsylvania and Illinois) account for 35% of the total schools on the list.

five_or_more = (states_nur > 3).sum()/len(states_nur)
print(five_or_more)
value_freq = states_nur.value_counts()
print(value_freq)

ax = value_freq.plot(kind='bar')
ax.set_xlabel('Number of Schools')
ax.set_ylabel('Frequency')

68% of the states have less than 4 schools on the top 230 list

Schools in the top 50 and where they are located

nur_top50 = nur_data.head(50)
nur_top50 = nur_top50['state_full'].value_counts(ascending=True)

ax = nur_top50.plot(kind='barh')
ax.set_xlabel('Number of Schools')
ax.set_ylabel('Frequency')

Schools in California, Massachusetts and New York account for close to half of the top 50 schools in the US

Checking to see if there is any relationship between the rank column and other numerical columns.

nur_data.corr()

The result shows the following:

1. A strong negative relationship between the rank and tuition fees columns

2. A moderate negative relationship between in-state and tuition fees

3. A moderate negative relationship between undergraduate enrollment and tuition fees

4. A moderate positive relationship between undergraduate enrollment and in-state

Visualizing the rank and tuition fees columns correlation

import matplotlib.pyplot as plt
%matplotlib inline
plt.scatter(nur_data['rank'], nur_data['tuition_and_fees'])

Visualizing the rank and tuition fees columns correlation

plt.scatter(nur_data['in_state'], nur_data['tuition_and_fees'])

Visualizing the undergraduate enrollment and tuition fees columns correlation

plt.scatter(nur_data['undergrad_enrollment'], nur_data['tuition_and_fees'])

plt.scatter(nur_data['undergrad_enrollment'], nur_data['in_state'])

States with the most expensive schools

mean_nur = nur_data.pivot_table(values='tuition_and_fees', index='state_full').sort_values(by='tuition_and_fees', ascending=True)

ax = mean_nur.plot(kind='barh', figsize=(20, 15))
ax.set_xlabel('Average Costs', fontsize= 30)
ax.set_ylabel('States', fontsize= 30)

The top five states with the most expensive schools are District of Columbia, New Hampshire, Connecticut, Massachusetts and Vermont.

The five least expensive schools are in South Dakota, Wyoming, Utah, North Dakota and Mississippi

[TelRich]

This is a nice insight that we may use when writing the article.

But, i don't think you've solved the task that was assigned to you, here. I don't see the code for that.

[TelRich]

Your task is as stated below:

Filter the data frame for the following 5 states. CA, NC, PA, MD, and NH. For each of them, there should be a dataframe sorted by tuition fees, in state, and undergrad enrollment.

I expect to see a code that outputs three data frames for each state. Let me know if you're having difficulties.

[Mueeshassan]

ca_state = nur_data[nur_data['state'] == 'CA']
ca_state_tuition = ca_state.sort_values(by='tuition_and_fees', ascending=False)
ca_state_instate = ca_state.sort_values(by='in_state', ascending=False)
ca_state_enrollment = ca_state.sort_values(by='undergrad_enrollment', ascending=False)

nc_state = nur_data[nur_data['state'] == 'NC']
nc_state_tuition = nc_state.sort_values(by='tuition_and_fees', ascending=False)
nc_state_instate = nc_state.sort_values(by='in_state', ascending=False)
nc_state_enrollment = nc_state.sort_values(by='undergrad_enrollment', ascending=False)

pa_state = nur_data[nur_data['state'] == 'PA']
pa_state_tuition = pa_state.sort_values(by='tuition_and_fees', ascending=False)
pa_state_instate = pa_state.sort_values(by='in_state', ascending=False)
pa_state_enrollment = pa_state.sort_values(by='undergrad_enrollment', ascending=False)

md_state = nur_data[nur_data['state'] == 'MD']
md_state_tuition = md_state.sort_values(by='tuition_and_fees', ascending=False)
md_state_instate = md_state.sort_values(by='in_state', ascending=False)
md_state_enrollment = md_state.sort_values(by='undergrad_enrollment', ascending=False)

nh_state = nur_data[nur_data['state'] == 'NH']
nh_state_tuition = nh_state.sort_values(by='tuition_and_fees', ascending=False)
nh_state_instate = nh_state.sort_values(by='in_state', ascending=False)
nh_state_enrollment = nh_state.sort_values(by='undergrad_enrollment', ascending=False)

[Mueeshassan]

I am sorry i did the wrong thing. I did not know that there was an instruction for the task. Please let me know if the updated code is good to go

[TelRich]

It's fine @Mueeshassan . I'll implement the new code