Read CSV exported from Calendly to create a bar chart with matplotlib depicting how students heard about program when booking an appointment. This version is more concise using the Counter class as central data structure.

lead_analyzer2.py

import csv
from collections import Counter
import matplotlib.pyplot as plt

# set up empty containers
leads = Counter()
email_addresses = set()
total_rows = 0 # storing total rows in csv if useful outside of plot

# ignores 'Other' entries; 'Other' since removed from appt booking question
how_students_heard_set = {"word of mouth", "green river website",
                          "greenriver.edu", "career event",
                          "internet search", "IT instructor",
                          "CS instructor", "social media",
                          "email campaign", "partner college referral",
                          "advising day"}

# iterate over rows in CSV file
with open("events-export.csv") as f:
    csv_reader = csv.DictReader(f)
    for row in csv_reader:
        total_rows += 1
        # focus on unique email addresses / students only
        if row["Invitee Email"] not in email_addresses:
            email_addresses.add(row["Invitee Email"])
            if row["Response 1"] in how_students_heard_set:
                leads[row["Response 1"]] += 1

# this bit of tidying up is needed due to Calendly question change from 'website' to '.edu'
leads["green river website"] += leads["greenriver.edu"]
del leads["greenriver.edu"]

# create sorted dict from Counter, by value, descending
sorted_leads = dict(leads.most_common())

# calculate totals for printing out
total = sum(v for v in sorted_leads.values())
for k,v in sorted_leads.items():
    print(f"{k} = {v} ({round((v/total*100), 1)}%)")

# extract leads data into two separate lists
sources = [key for key in sorted_leads.keys()]
values = [value for value in sorted_leads.values()]

# set up fig and axs
plt.style.use("seaborn")
fig, ax = plt.subplots()
ind = range(len(values))
plot = ax.bar(ind, values, color="#28a745")

# display data value on head of each bar
for rect in plot:
    height = rect.get_height()
    ax.text(rect.get_x() + rect.get_width() / 2, 1.002 * height,
            f"{int(height)}", ha="center", va="bottom")

# configure ticks and lims
ax.set_xticks(ind) # source for source in sources
ax.set_xticklabels(sources)
ax.set_ylim(0, 180)

# configure labels, title, and annotations
ax.set_ylabel("Unique Responses")
plt.title("How students report hearing about our BAS-SD: July 2017 - PRESENT",
          size="large", weight="bold")
plt.annotate("Data collected at booking via Calendly", xy=(8.105, 175.000))
plt.annotate(f"{total} unique student data points", xy=(8.450, 170.500))
plt.annotate("Plot by Andy O", xy=(9.150, 166.000))

# plot!
plt.show()