2.2: Create a vocabulary profile

During this step I will make use of a convenient internet tool, called the VocabProfile (VP). This program performs a lexical text analysis and divides the words into four catagories:
1. K1: The most frequent 1000 words of English
2. K2: The second most frequent 1000 words of English
3. AWL: The 570 most frequently used words in academic texts
4. Off-list: The remainder which are not found in categories 1,2 or 3

By looking at the results from this analysis, I can improve my text by using different words. This is the result from the analysis of the initial text:

WEB VP OUTPUT FOR FILE: Day Chocolate

Words recategorized by user as 1k items (proper nouns etc): NONE (total 0 tokens)

Families Types Tokens Percent K1 Words (1-1000): 105 128 291 78.23%   Function: ... ... (159) (42.74%)   Content: ... ... (132) (35.48%) >   Anglo-Sax       =Not Greco-Lat/Fr Cog: ... ... (82) (22.04%) K2 Words (1001-2000): 9 10 18 4.84% >   Anglo-Sax:      ... ... (2) (0.54%)     1k+2k       ... ... (83.07%) AWL Words (academic): 26 26 31 8.33% >   Anglo-Sax:      ... ... (1) (0.27%) Off-List Words: ? 19 32 8.60% 140+? 183 372 100%

Words in text (tokens): 372 Different words (types): 183 Type-token ratio: 0.49 Tokens per type: 2.03 Lex density (content words/total) 0.57 Pertaining to onlist only Tokens: 340 Types: 164 Families: 140 Tokens per family: 2.43 Types per family: 1.17 Anglo-Sax Index: (A-Sax tokens + functors / onlist tokens) 71.76% Greco-Lat/Fr-Cognate Index: (Inverse of above) 28.24%

When comparing my results, it is clearly visible that I am not a native speaker.The results from an educated native speaker of English would be 70% on K1 and 10% on each of the remaining categories.

Now it is time to upgrade the text to make it more academical. This can be done by looking at another function of the VP: The Token List.

The Toke List provides me with an overview of what words I used and from what sublist these words are. The following words are the AWL words that I used in my text.

Sublist 1
created environmentally establish establish major responsiveness

---

Sublist 2
achieve assists consuming maintain normal normal obtain resources strategy strategy

---

Sublist 3
alternatives considerable ensure technique

---

Sublist 5
aware enabled networks

---

Sublist 6
bond bond initiated

---

Sublist 7
channel

---

Sublist 8
eventually

---

Sublist 9
diminishes ethically ethically

As we can see, there are almost no AWL words that I used more than once. Therefore, it is not necessary to look if there are other words that I could have used. But, during the self-editing steps, I noticed that I could have used some AWL words instead of words from the K1 or K2 category. When changing these words the following vocabulary profile results:

WEB VP OUTPUT FOR FILE: Day Chocolate

Words recategorized by user as 1k items (proper nouns etc): NONE (total 0 tokens)

Families Types Tokens Percent K1 Words (1-1000): 102 125 321 75.71%   Function: ... ... (179) (42.22%)   Content: ... ... (142) (33.49%) >   Anglo-Sax       =Not Greco-Lat/Fr Cog: ... ... (89) (20.99%) K2 Words (1001-2000): 10 12 22 5.19% >   Anglo-Sax:      ... ... (3) (0.71%)     1k+2k       ... ... (80.90%) AWL Words (academic): 34 38 46 10.85% >   Anglo-Sax:      ... ... (3) (0.71%) Off-List Words: ? 18 35 8.25% 146+? 193 424 100%

Words in text (tokens): 424 Different words (types): 193 Type-token ratio: 0.46 Tokens per type: 2.20 Lex density (content words/total) 0.58 Pertaining to onlist only Tokens: 389 Types: 175 Families: 146 Tokens per family: 2.66 Types per family: 1.20 Anglo-Sax Index: (A-Sax tokens + functors / onlist tokens) 70.44% Greco-Lat/Fr-Cognate Index: (Inverse of above) 29.56%

The profile shows that the percentage of AWL words increased form 8.33 percent to 10.85 percent and hereby increasing the academic value of this text.

Reflection:
This tool was absolutely useful, mainly due to its easy interface and clear structure. It is especially for someone like me, someone who has trouble implementing AWL words, a very helpful tool.