WordStat Dictionaries

Brand Personality Dictionary

This dictionary uses Aaker's five brand personality dimensions (Sincerity, Excitement, Competence, Sophistication and Ruggedness). It lists 833 words across these five dimensions.

For example: There are 174 word categorised as Sincerity. These include:

• Candid
• Frank
• Good
• Merciful
• Kind, etc...

WordStat Sentiment Dictionary

The WordStat Sentiment Dictionary was designed by combining negative and positive words from the Harvard IV dictionary, the Regressive Imagery Dictionary (Martindale, 2003), and the Linguistic and Word Count dictionary (Pennebaker, 2007). It was expanded by automatically identifying potential synonyms and related words as well as any inflected forms. There are more than 9526 negative and 4669 positive word patterns. Sentiment is not measured by the two word list but by two rules that attempt to take into account negations that may precede those words. For example, negative sentiment is measured by using the following two rules:

1. Negative words not preceded by a negation (no, not never) within four words in the same sentence.
2. Positive words preceded by a negation within four words in the same sentence.

Positive sentiment is measured in a similar way by looking for positive words not preceded by a negation as well as negative terms following a negation. Care: Provalis Research experience suggests that this last rule has less predictive value and may even slightly deteriorate the measurement of sentiment. There may be some situations where such a rule could help predict positive sentiment. Provalis Research kept this last rule so you can decide whether it should be applied or disabled.

To use the dictionary in WordStat, extract it in the My Provalis Research Projects\Dictionaries folder located in your main Documents folder.

PROVALIS RESEARCH DO NOT RECOMMEND USING THIS DICTIONARY AS IS. They strongly believe that doing so will not present very accurate results. They recommend instead customising this dictionary by removing domain specific words (i.e. words not associated with your domain of research), indentifying incorrect predictions and adding domain specific words for your domain of research.

The Loughran and McDonald Financial Sentiment Dictionary

The Loughran and McDonald (2011) article provides a clear demonstration that applying a general sentiment word list to accounting and finance topics can lead to a high rate of misclassification. They found that about three-quarters of the negative words in the Harvard IV TagNeg dictionary of negative words are typically not negative in a financial context. For example, words like “mine”, “cancer”, “tire” or “capital” are often used to refer to a specific industry segment. These words are not predictive of the tone of documents or of financial news and simply add noise to the measurement of sentiment and attenuate its predictive value. These authors created custom lists of negative and positive words specific to the accounting and financial domain. Another benefit of the dictionary they propose is that it shows how quantitative content analysis can move beyond mere dichotomous differentiation typical of sentiment analysis and can also be used to measure additional dimensions of interest. Two noteworthy additions are the Uncertainty word list that attempts to measure the general notion of imprecision (without an explicit reference to risks), and the Litigiousness word list that may be used to identify potential legal problem situations. They also included Weak Modal and Strong Modal word lists. The following table illustrates the various categories of the Loughran and McDonald financial sentiment dictionary.

Regressive Imagery Dictionary – By Colin Martindale

The English Regressive Imagery Dictionary (RID) is composed of about 3200 words and roots assigned to 29 categories of primary process cognition, 7 categories of secondary process cognition, and 7 categories of emotions.

The Regressive Imagery Dictionary (Martindale, 1975,1990) is a content analysis coding scheme designed to measure primordial vs. conceptual thinking. Conceptual thought is abstract, logical, reality oriented, and aimed at problem-solving. Primordial thought is associative, concrete, and takes little account of reality. It is the type of thinking found in fantasy, reverie, and dreams. A running tally of category occurrences is kept, and the final output consists of the categories and the percentage of words in a document that were assigned to each category. This output can then be subjected to statistical analyses. These categories were derived from the theoretical and empirical literature on regressive thought (Martindale, 1975, 1990). The rationale behind the dictionary is that psychological processes will be reflected in the content of a text. Thus, for example, the more primordial the thought involved in producing a text, the less abstract and the more drive- and sensation-oriented words it should contain. It should be noted that the view of regressive cognition upon which construction of the dictionary was based is not a narrowly psychoanalytic one but is closer to Werner’s (1948) or Goldstein’s views (1939). Thus, the construct that the dictionary ultimately measures might as well be called dedifferentiated thinking as regressive or primary process thinking. For these reasons, we use the terms conceptual vs. primordial thought.

This dictionary has primary processes, secondary processes and emotions. An example of a primary process category is: Sensation - which has sub categories of Touch, Odour etc. The sub category Touch has words: Touch, thick, and stroke.

An example of a secondary process is: Restraint - which has words: Must, stop, bind.

An example of an emotion is: Anxiety - which is associated with Afraid, fear, phobic.

Forest Value Dictionary

The Forest Value Dictionary is composed of 612 words and phrases. It is a classification system that identifies four broad categories of forest values: economic/utilitarian, life support, aesthetic, and moral/spiritual values. The dictionary can be modified for use with other subjects. In 2017 it was modified for a study on marine issues by Peter J. Jacques and Rafaella Lobo, SOFIA’S choices: Discourses, values, and norms of the World Ocean Regime. It was originally developed and used to empirically analyse the evolution of forest values in the United States from 1982 through 1993 (Bengston & Xu, 1995).

If you need further information on the Forest Value Dictionary, please contact David N. Bengston at dbengston@fs.fed.us or visit his web site.

An example from the dictionary is the category Life Support Value Expressions, which is linked to "absorption of pollutants, acid rain, air quality, biodiversity, carbon storage, climate change, damaging steam, ecosystem, erosion, extinct species, habitat, ozone layer".

Corporate Social Responsibility – By Pencle and Malaescu (2016)

The Corporate Social Responsibility content analytic dictionary was developed by Nadra Pencle and Irina Malaescu (2016) and used to predict various components of Initial Public Offering (IPO) pricing. The authors analysed the prospectus of U.S. IPOs from 2011-2013, in part, because the IPO process has the potential to convey usable information that can signal business plans and managers’ level of confidence, as well as potential problems in their businesses (Loughran and MacDonald 2016). Therefore, it can be reasoned that despite the lack of regulatory requirements, a company entering into the public equity market and wishing to signal to potential investors its stance on CSR orientation may do so using a more detailed CSR disclosure in its prospectus than it might otherwise. Pencle and Maleaescu (2016) developed a multi-dimensional corporate social responsibility words (CSRW) dictionary with four dimensions; Social and Community, Human Resources, Environment, and Human Rights.

In line with the guidelines by Short, Broberg, Cogliser, and Brigham (2010) and Kenny et al (2013) both a deductive and inductive wordlist was generated using related CSR literature. With the help of qualitative analysis and content analysis software tools, the authors produced an inductive wordlist of 1,002 words. To validate the final wordlists three judges with significant expertise in the field of CSR were enlisted. The final dictionary is comprised of words on which consensus was reached.

Communication vagueness dictionary by J.H. Hiller

Vagueness is a psychological construct which refers to the ‘state of mind’ of a communicator who does not sufficiently command the facts, knowledge, or understanding required for maximally effective communication. Vagueness is an internal stimulus condition principally determined by a performer’s instant command of knowledge, by his motivation to communicate, and by his own evaluation of imperfect success. Use of the words and phrases in the Vagueness dictionary provide the analyst with a clue to the communicator’s vague state of cognition to uncertainty of facts and/or understanding of the thought underlying the communication. So, for example, when a speaker hesitates and then says, Well, as “you all know,” and “you know,” or says, “clearly,” “obviously,” or “naturally,” the analyst may infer that most folks in the audience do not know, and the reasons are not clear to the speaker, so that the speaker has opted to bluff his way through the topic to be addressed.

Some examples:

Negated Intensifiers: not always, not quite, isn't necessarily.

Approximation: nearly, sort of, fairly, almost, pretty much.

This dictionary is sold by Provalis Research. Clicking on the "buy" link below will allow you to place an order directly on their website

Body Type Dictionary

The Body Type Dictionary is an attempt to operationalise Fisher and Cleveland’s (1958) theory of body-boundary strength for use in computer-assisted content analysis. Working originally with Rorschach ink-blot protocols, Fisher and Cleveland identified two broad personality types: “High Barrier” individuals, who have a confident sense of their own self boundaries, and “Low Barrier” individuals, who have a less clear sense of their distinct identity and are inclined towards what one might call a “de-differentiated” image of themselves in relation to others and the outside world in general. In turn, Fisher and Cleveland (and subsequent researchers) have linked these personality types to a wide range of other personality, health, and social variables. Fisher and Cleveland’s person categorisations were based on two kinds of subconscious verbal imagery in their Rorschach protocols: barrier imagery, which shows a concern for protective boundaries, and penetration imagery, which relates to the destruction or penetration of such boundaries. In their 1958 book, they provide a detailed set of manual coding instructions and sample coded Rorschach protocols, on which the Body Type Dictionary is based. The dictionary contains 515 words stored in 12 categories.

When using this dictionary for articles to be published, all authors must cite:

Wilson, A. (2006). Development and application of a content analysis dictionary for body boundary research. Literary and Linguistic Computing, 21, 105-110.

Laver & Garry Dictionary of Policy Position

The Laver & Garry dictionary was developed to estimate the policy positions of political actors in the United Kingdom by comparing their speeches and written documents to keywords found in the British conservative and Labour manifestos of 1992. Words have been selected: semantically based on their relation to specific content categories; and empirically based on specificity to a particular political party (a word should be used by one party at least twice as often as by the other party). The dictionary contains 415 words and word patterns stored in 19 categories.

Note: This dictionary was customised to reflect the policy positions of UK political parties. It has limited utility to assess political documents from other countries. The procedure outline in Laver & Gary (2000) could, however, be used to adapt the dictionary to a different political context.

When using this dictionary for articles to be published, all authors must cite:

Laver. M. & Garry, J. (2000). Estimating Policy Positions from Political Texts. American Journal of Political Science, 44 (3), 619-634.

WordNet-based Categorisation Dictionary

This categorisation dictionary is derived from the WordNet® database to provides basic categorisations of noun, verbs, adjectives and adverbs currently found in the WordNet 2.0 database into 44 syntactic category and logical groupings. Four versions of this categorisation dictionary are currently available:

1. Words only
• The full version offers categorisation of 109231 words into 44 WordNet lexical categories. More than 43,000 of those entries are categorised in more than one category.
• The limited version consists of 65425 unambiguous words categorised into those same categories. Unambiguous words are defined as words that are categorised in only one Wordnet lexical category.

1. Words and phrases
• The full version offers categorisation of 174268 words and phrases into 44 WordNet lexical categories. More than 47,000 of those entries are categorised in more than one category.
• The limited version consists of 126869 unambiguous words and phrases categorised into those same categories. Unambiguous words and phrases are defined as entries that are categorised in only one WordNet lexical category.

Note: WordNet® is an online lexical reference system developed by the Cognitive Science Laboratory at Princeton University under the direction of Professor George A. Miller (Principal Investigator). WordNet 1.7.1 Copyright © 2001 by Princeton University. All rights reserved.

Roget's Categorisation Dictionary

This WordStat dictionary based on the well-known Roget's Thesaurus. It categorises 100,685 words and phrases into 1042 categories (6 broad classes).

This dictionary was created using the 1911 edition of Roget's Thesaurus. Many modern words are missing. For this reason, it may not be entirely suitable for use in natural language processing tasks. It may nevertheless be useful for experimentation as well as for the development of more comprehensive categorisation systems.