diff --git a/python/data science/poop, klutzy.csv b/python/data science/poop, klutzy.csv
new file mode 100644
index 0000000..f83dfed
--- /dev/null
+++ b/python/data science/poop, klutzy.csv	
@@ -0,0 +1,3 @@
+,word,stdev,mean,median,mode,range,q1,q3,iqr
+0,poop,2.599393484092581e-07,6.139207233218996e-07,6.637646704023479e-07,9.126888329547e-07,1.069131408420227e-06,4.1074716961020385e-07,8.147061285918815e-07,4.039589589816777e-07
+1,klutzy,7.003951820605788e-09,3.2869637653755984e-09,0.0,0.0,2.649079254370333e-08,0.0,1.0566731666248965e-10,1.0566731666248965e-10
diff --git a/python/data science/poop, klutzy.png b/python/data science/poop, klutzy.png
new file mode 100644
index 0000000..aeb2fb5
Binary files /dev/null and b/python/data science/poop, klutzy.png differ
diff --git a/python/data science/wordPopularity.py b/python/data science/wordPopularity.py
new file mode 100644
index 0000000..7785a1d
--- /dev/null
+++ b/python/data science/wordPopularity.py	
@@ -0,0 +1,69 @@
+import statistics
+import requests
+import json
+import matplotlib.pyplot as plt
+import numpy
+import os
+import pandas
+
+pandas.options.display.float_format = '{:.10f}'.format
+
+words = input('Please enter a list of words. Separate each word with a comma (:\n')
+print()
+startYear = 1800
+endYear = 2019
+years = range(startYear, endYear + 1)
+
+response = requests.get('https://books.google.com/ngrams/json?content=%s&year_start=%s&year_end=%s&corpus=26&smoothing=3' % (words, startYear, endYear))
+data = json.loads(response.content)
+
+frames = []
+
+for x in data:
+	frame = {}
+	frame['word'] = x['ngram']
+	frame['stdev'] = numpy.std(x['timeseries'])
+	frame['mean'] = numpy.mean(x['timeseries'])
+	frame['median'] = numpy.median(x['timeseries'])
+	frame['mode'] = statistics.mode(x['timeseries'])
+	frame['range'] = max(x['timeseries']) - min(x['timeseries'])
+	frame['q1'] = numpy.percentile(x['timeseries'], 25)
+	frame['q3'] = numpy.percentile(x['timeseries'], 75)
+	frame['iqr'] = frame['q3'] - frame['q1']
+	frames.append(frame)
+	plt.plot(years, x['timeseries'], label=frame['word'])
+
+df = pandas.DataFrame(frames)
+print(df)
+
+# Summary Statistics
+for frame in frames:
+	print()
+	if frame['stdev'] == max([ f['stdev'] for f in frames ]):
+		print('%s has the highest standard deviation!' % (frame['word']))
+	elif frame['stdev'] == min([ f['stdev'] for f in frames ]):
+		print('%s has the lowest standard deviation!' % (frame['word']))
+
+# Save CSV
+dirName = os.path.dirname(os.path.realpath(__file__))
+wordList = ', '.join([f['word'] for f in frames])
+
+while True:
+	toSave = input('Would you like to save this data frame in a CSV? (y/n)')
+	if toSave == 'y':
+		df.to_csv('%s/%s.csv' % (dirName, wordList))
+		break
+	if toSave == 'n':
+		break
+
+# Save Graph
+while True:
+	toSave = input('Would you like to save a graph of the data? (y/n)').lower()
+	if toSave == 'y':
+		plt.ticklabel_format(style='plain')
+		plt.legend()
+		plt.savefig('%s/%s.png' % (dirName, wordList), dpi=100)
+		exit()
+	elif toSave == 'n':
+		exit()
+
diff --git a/readme.md b/readme.md
index 1f5ed31..cda93cf 100644
--- a/readme.md
+++ b/readme.md
@@ -12,6 +12,7 @@
 - [Data Science](python/data%20science)
   - [Fake Jobs Scraper](python/data%20science/fakejobs.py)
   - [Country Population vs Active Facebook Users in the Country](python/data%20science/fbPercentActive.py)
+  - [ngrams Word Popularity](python/data%20science/wordPopularity.py)
 - [Calculators](python/calculators)
   - [Binomial Distribution](python/calculators/Binomial%20Distribution.py)
   - [Pearson's Product-Moment Correlation Coefficient](python/calculators/PMCC.py)