Word Analysis of 2012 U.S. Presidential Debates

Barack Obama vs Mitt Romney (2nd debate)

16 October 2012

Introduction

Word Statistics

Debate Word Count

Summary Word Count

The summary word count reports the total number of words and the number of unique, non-stop words used by each candidate. Word number is expressed as both absolute and relative values.

Table 1a

all words

Number of all words and unique words used by each speaker.

set

word count

Barack Obama

7,592 download text •	1,308
48.5%	17.2%

Mitt Romney

8,062 download text •	1,233
51.5%	15.3%

total

15,654	1,894
100.0%	12.1%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 1b

exclusive and shared words

Words exclusive to speaker (e.g. speaker A but not speaker B) and shared by speakers (speaker A and B).

set

word count

Barack Obama

1,006 download text •	661
13.3%	65.7%

Mitt Romney

978 download text •	586
12.1%	59.9%

both candidates

13,670 download text •	647
87.3%	4.7%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 1

commentary

Table 1

legend

a	c
b	d

a :: word count

b :: word count, as fraction in total in debate

c :: unique words in (a)

d :: unique words in (a), as fraction in (a) bar :: proportion of (a-c):c

Stop Word Contribution

In the table below, the candidates' delivery is partitioned into stop and non-stop words. Stop words (full list) are frequently-used bridging words (e.g. pronouns and conjunctions) whose meaning depends entirely on context. The fraction of words that are stop words is one measure of the complexity of speech.

Table 2a

non-stop words

Counts of stop and non-stop words.

speaker

all

stop

non-stop

Barack Obama

7,592 download text •	1,308
100.0%	17.2%

4,243 download text •	146
55.9%	3.4%

3,349 download text •	1,162
44.1%	34.7%

Mitt Romney

8,062 download text •	1,233
100.0%	15.3%

4,526 download text •	144
56.1%	3.2%

3,536 download text •	1,089
43.9%	30.8%

total

15,654 download text •	1,894
100.0%	12.1%

8,769 download text •	158
56.0%	1.8%

6,885 download text •	1,736
44.0%	25.2%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 2b

exclusive and shared non-stop words

Non-stop words exclusive to speaker (e.g. speaker A but not speaker B) and shared by speakers (speaker A and B).

set

word count

Barack Obama

979 download text •	647
29.2%	66.1%

Mitt Romney

952 download text •	574
26.9%	60.3%

both candidates

4,954 download text •	515
72.0%	10.4%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 2

commentary

Table 2

legend

a	c
b	d

a :: total number of words, for a given category (all, stop, non-stop)

b :: (a) relative to words in the debate if category=all, otherwise relative to words by the candidate

c :: number of unique words with set (a)

d :: (c) relative to (a)

bar :: proportion of (a-c):c

All further word use statistics represent content that has been filtered for stop words.

Word frequency

The word frequency table summarizes the frequency with which words were used. I show the average word frequency and the weighted cumulative frequencies at 50 and 90 percentile. The average word frequency indicates how many times, on average, a word is used. For a given fraction of the entire delivery, the weighted cumulative frequency indicates the largest word frequency within this fraction (details about weighted cumulative distribution).

Table 3a

word use frequency

Average and 50%/90% percentile word frequencies.

speaker

word frequency

all

stop

non-stop

Barack Obama

5.8 download text •	23	271

29.1 download text •	84	275

2.9 download text •	5	32

Mitt Romney

6.5 download text •	25	263

31.4 download text •	92	310

3.2 download text •	6	30

total

8.3	47	512

55.5	167	596

4.0	9	53

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 3b

exclusive and shared non-stop word use frequency

Average and 50%/90% cumulative percentile word frequencies. Non-stop words exclusive to speaker (e.g. speaker A but not speaker B) and shared by speakers (speaker A and B).

set

word frequency

Barack Obama

1.51 download text •	2	5

Mitt Romney

1.66 download text •	2	6

total

3.97 download text •	9	53

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 3

commentary

Table 3

legend

a	b	c

a :: average word frequency

b :: largest word frequency in 50% of content

c :: largest word frequency in 90% of content

bar :: proportion of a:b:c

Sentence Size

Table 4

sentence size

Number of sentences spoken by each speaker and sentence word count statistics. Number of words in a sentence is shown by average and 50%/90% cumulative values for all, stop and non-stop words.

speaker

number of sentences

sentence size

all

stop

non-stop

Barack Obama

500

15.2 download text •	22	45

8.8 download text •	13	25

6.8 download text •	10	20

Mitt Romney

586

13.8 download text •	19	40

7.9 download text •	11	24

6.2 download text •	8	18

total

1,086

16.4	21	43

10.3	13	25

8.5	10	20

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 4

commentary

Table 4

legend

a	b	c

a :: average sentence size

b :: largest sentence size for 50% of content

c :: largest sentence size for 90% of content

bar :: proportion of a:b:c

Part of Speech Analysis

In this section, word frequency is broken down by their part of speech (POS). The four POS groups examined are nouns, verbs, adjectives and adverbs. Conjunctions and prepositions are not considered. The first category (n+v+adj+adv) is composed of all four POS groups.

Part of Speech Count

Table 5

part of speech count

Count of words categorized by part of speech (POS).

part of speech

n+v+adj+adv

nouns (n)

verbs (v)

adjectives (adj)

adverbs (adv)

Barack Obama

3,057 download text •	1,100
40.3%	36.0%

1,491 download text •	585
48.8%	39.2%

861 download text •	360
28.2%	41.8%

558 download text •	254
18.3%	45.5%

147 download text •	50
4.8%	34.0%

Mitt Romney

3,188 download text •	1,039
39.5%	32.6%

1,565 download text •	564
49.1%	36.0%

882 download text •	331
27.7%	37.5%

588 download text •	263
18.4%	44.7%

153 download text •	59
4.8%	38.6%

total

6,245 download text •	1,655
39.9%	26.5%

3,056 download text •	901
48.9%	29.5%

1,743 download text •	548
27.9%	31.4%

1,146 download text •	427
18.4%	37.3%

300 download text •	84
4.8%	28.0%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 5

commentary

Table 5

legend

a	c
b	d

a :: total number of words for a given POS (all, noun, verb, adjective, adverb, pronoun)

b :: (a) relative to all words by candidate

c :: unique words in (a)

d :: (c) relative to (a)

bar :: proportion of (a-c):c

Part of Speech Frequency

Table 5

part of speech frequency

Frequency of words categorized by part of speech (POS).

part of speech frequency

n+v+adj+adv

nouns (n)

verbs (v)

adjectives (adj)

adverbs (adv)

Barack Obama

2.78	5	28

2.55	4	24

2.39	4	34

2.20	3	16

2.94	7	32

Mitt Romney

3.07	6	24

2.77	5	30

2.67	5	26

2.24	3	16

2.59	3	21

total

3.77	9	45

3.39	8	45

3.18	6	42

2.68	5	24

3.57	6	53

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 5

commentary

Table 5

legend

a	b	c

a :: average word frequency

b :: largest word frequency in 50% of content

c :: largest word frequency in 90% of content

bar :: proportion of a:b:c

Part of Speech Pairing

Through word pairing, I extract concepts from the text. The number of unique word pairs is a function of sentence length and is one of the measures of complexity.

Table 6a

part of speech pairing — Barack Obama

Word pairs (total and unique) categorized by part of speech (POS)

part of speech pairings - Barack Obama

noun

verb

adjective

adverb

noun

3,667 download text •	3,067
	83.6%

verb

4,297 download text •	3,750
	87.3%

1,090 download text •	966
	88.6%

adjective

2,287 download text •	1,955
	85.5%

1,318 download text •	1,102
	83.6%

386 download text •	330
	85.5%

adverb

716 download text •	609
	85.1%

389 download text •	346
	88.9%

189 download text •	166
	87.8%

28 download text •	25
	89.3%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 6b

part of speech pairing — Mitt Romney

Word pairs (total and unique) categorized by part of speech (POS)

part of speech pairings - Mitt Romney

noun

verb

adjective

adverb

noun

3,342 download text •	2,772
	82.9%

verb

3,616 download text •	3,099
	85.7%

944 download text •	798
	84.5%

adjective

2,233 download text •	1,874
	83.9%

1,212 download text •	1,012
	83.5%

328 download text •	279
	85.1%

adverb

603 download text •	533
	88.4%

359 download text •	309
	86.1%

213 download text •	191
	89.7%

39 download text •	35
	89.7%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 6c

unique part of speech pairing — candidate comparison

Unique word pairs categorized by part of speech (POS)

unique part of speech pairings

noun (n)

verb (v)

adjective (adj)

adverb (adv)

noun

3,067 download text •	2,772 download text •
	90.4%

verb

3,750 download text •	3,099 download text •
	82.6%

966 download text •	798 download text •
	82.6%

adjective

1,955 download text •	1,874 download text •
	95.9%

1,102 download text •	1,012 download text •
	91.8%

330 download text •	279 download text •
	84.5%

adverb

609 download text •	533 download text •
	87.5%

346 download text •	309 download text •
	89.3%

166 download text •	191 download text •
	115.1%

25 download text •	35 download text •
	140.0%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 6

commentary

Table 6 a,b

legend

a	c
	d

a :: total number of pairs, for a given category (e.g. verb/noun)

c :: number of unique pairs within set (a)

d :: (c) relative to (a)

bar :: proportion of (a-c):c

Table 6c

legend

a	c
	d

a :: unique pairs for Barack Obama

c :: unique pairs for Mitt Romney

d :: (c) relative to (a) (i.e. Mitt Romney relative to Barack Obama)

bars :: (a) and (c)

Exclusive and Shared Usage

This section enumerates words that were exclusive to a candidate (e.g. used by one candidate but not the other). This content provides insight into what the candidates' priorities are and reveals differences in perspective on similar topics.

For a given part of speech, the table breaks down the number of words that were spoken by only one of the candidates or both candidates (intersection). The last row includes words spoken by either candidate (union).

Table 7

exclusive word usage

Total and unique words used exclusively by a candidate, or by both.

part of speech

n+v+adj+adv

nouns (n)

verbs (v)

adjectives (adj)

adverbs (adv)

Barack Obama

928 download text •	616
100.0%	66.4%
14.9%	37.2%

455 download text •	318
49.0%	69.9%
14.9%	35.3%

273 download text •	190
29.4%	69.6%
15.7%	34.7%

168 download text •	126
18.1%	75.0%
14.7%	29.5%

32 download text •	25
3.4%	78.1%
10.7%	29.8%

Mitt Romney

925 download text •	555
100.0%	60.0%
14.8%	33.5%

418 download text •	282
45.2%	67.5%
13.7%	31.3%

249 download text •	157
26.9%	63.1%
14.3%	28.6%

211 download text •	139
22.8%	65.9%
18.4%	32.6%

47 download text •	29
5.1%	61.7%
15.7%	34.5%

both candidates

4,392 download text •	484
100.0%	11.0%
70.3%	29.2%

2,087 download text •	248
47.5%	11.9%
68.3%	27.5%

1,145 download text •	143
26.1%	12.5%
65.7%	26.1%

659 download text •	90
15.0%	13.7%
57.5%	21.1%

211 download text •	25
4.8%	11.8%
70.3%	29.8%

total

6,245 download text •	1,655
100.0%	26.5%
100.0%	100.0%

3,056 download text •	901
48.9%	29.5%
100.0%	100.0%

1,743 download text •	548
27.9%	31.4%
100.0%	100.0%

1,146 download text •	427
18.4%	37.3%
100.0%	100.0%

300 download text •	84
4.8%	28.0%
100.0%	100.0%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 7

commentary

Table 7c

legend

a	d
b	e
c	f

a :: total number of words in set (e.g. obama \ romney, obama ∩ romney, obama ∪ romney , for a given part of speech

b :: (a) relative to all exclusive words in n+v+adj+adv

c :: (a) relative to all words in n+v+adj+adv

d :: unique words in (a)

e :: (d) relative to (a)

f :: (d) relative to all unique words in n+v+adj+adv

bar1 :: normalized ratio of (a-d):d

bar2 :: absolute ratio of (a-d):d for all POS groups (first column) or POS group (other columns)

Noun Phrase Usage

Noun phrases were extracted from the text and analyzed for frequency, word count, unique word count and richness. Single-word phrases were not counted.

Top-level noun phrases are those without a parent noun phrase (a parent phrase is one that a similar, longer phrase). Derived noun phrases are those with a parent (more details about noun phrase analysis).

The top-level noun phrases can be interpreted as independent concepts. Derived noun phrases can be interpreted as variants on concepts embodied by the top-level phrases.

Noun Phrase Count and length

This table reports the absolute number of noun phrases, which is related to the number of nouns, and their length.

Table 8a

noun phrase count

Counts of noun phrases in words and per noun.

speaker

noun phrase count

all

top-level

Barack Obama

529 download text •	240
100.0%	45.4%
0.35	0.41

433 download text •	234
81.9%	54.0%
0.29	0.40

Mitt Romney

549 download text •	241
100.0%	43.9%
0.35	0.43

439 download text •	231
80.0%	52.6%
0.28	0.41

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 8b

noun phrase length

Average and 50%/90% cumulative length of noun phrases, in words.

speaker

noun phrase length

all

top-level

Barack Obama

2.31	2	3

2.36	2	4

Mitt Romney

2.29	2	3

2.35	2	4

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 8

commentary

Table 8a

legend

a	d
b	e
c	f

a :: number of noun phrases

b :: (a) relative to number of all noun phrases

c :: number of noun phrases per noun

d :: number of unique phrases

e :: (c) relative to (a)

f :: number of unique noun phrases per unique noun

bar :: normalized ratio of (a-c):c

Table 8b

legend

a	b	c

a :: average noun phrase size, in words

b :: largest noun phrase size in 50% of content

c :: largest noun phrase size in 90% of content

bar :: proportion of a:b:c

Exclusive and Shared Noun Phrase Count and length

Table 9a

exclusive and shared noun phrase count

Counts of exclusive and shared noun phrases in words and per noun.

speaker

noun phrase count

all

top-level

Barack Obama

448 download text •	225
41.6%	50.2%

399 download text •	226
89.1%	56.6%

Mitt Romney

458 download text •	231
42.5%	50.4%

399 download text •	224
87.1%	56.1%

both candidates

172 download text •	31
16.0%	18.0%

74 download text •	18
43.0%	24.3%

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 9b

exclusive and shared noun phrase length

Average and 50%/90% cumulative length of noun phrases, in words.

speaker

noun phrase length

all

top-level

Barack Obama

2.35	2	4

2.38	2	4

Mitt Romney

2.34	2	4

2.36	2	4

both candidates

2.08	2	3

2.18	2	3

Fields with • (e.g. 155•) link to data files. Hover over the field to show these links.

Table 9

commentary

Table 9a

legend

a	c
b	d

a :: number of noun phrases

b :: (a) relative to number of all noun phrases

c :: number of unique phrases

d :: (c) relative to (a)

bar :: normalized ratio of (a-c):c

Table 9b

legend

a	b	c

a :: average noun phrase size, in words

b :: largest noun phrase size in 50% of content

c :: largest noun phrase size in 90% of content

bar :: proportion of a:b:c

Windbag Index

The Windbag Index is a compound measure that characterizes the complexity of speech. A low index is indicative of succinct speech with low degree of repetition and large number of independent concepts.

Table 10

windbag index

Windbag Index for each speaker. The higher the value, the more repetitive the speech.

speaker

Windbag Index

index value

index terms

Barack Obama

581
-22.9%

0.441	0.347	0.392	0.418	0.455	0.340	0.454	0.975
+0.6%	+12.7%	+8.9%	+11.4%	+1.8%	-11.8%	+3.4%	+1.7%

Mitt Romney

754
+29.7%

0.439	0.308	0.360	0.375	0.447	0.386	0.439	0.959
-0.6%	-11.2%	-8.1%	-10.2%	-1.7%	+13.4%	-3.2%	-1.7%

Table 10

commentary

Table 10

legend

The Windbag Index is 1/(t1*t2*...*t9) where t1,t2,...,t8 are

t1 :: fraction of words which are non-stop

t2 :: fraction of non-stop words which are unique

t3 :: fraction of nouns which are unique

t4 :: fraction of verbs which are unique

t5 :: fraction of adjectives which are unique

t6 :: fraction of adverbs which are unique

t7 :: fraction of noun phrases which are unique

t8 :: fraction of noun phrases which are top-level

Large individual terms t1...t9 contribute to a smaller index.

The percentage values below the index and each term are relative differences to the other speaker's corresponding term (i.e. 100*(a-b)/b where a is the value for one speaker and b for the other).

Word Clouds

In the word clouds below, the size of the word is proportional to the number of times it was used by a candidate (method details).

Not all words from a group used to draw the cloud fit in the image — less frequently used words for large word groups may fall outside the image.

All Words for Each Candidate

Each candidate's debate portion was extracted and frequencies were compiled for each part of speech (noun, verb, adjective, adverb), with words colored by their part of speech category.

The distribution of sizes within a tag cloud follows the frequency distribution of words. However, word size cannot be compared between clouds, since the minimum and maximum size of the words is fixed.

Debate Word Cloud for Barack Obama - all words

Debate Word Cloud for Mitt Romney - all words

commentary

Exclusive Words for Each Candidate

The clouds below show words used exlusively by a candidate. For example, if candidate A used the word "invest" (any number of times), but candidate B did not, then the word will appear in the exclusive word tag cloud for candidate A.

Words exclusive to Barack Obama

Words exclusive to Mitt Romney

commentary

Part of Speech Word Clouds

In these clouds, words from each major part of speech were colored based on whether they were exclusive to a candidate or shared by the candidates.

The size of the word is relative to the frequency for the candidate — word sizes between candidates should not be used to indicate difference in absolute frequency.