T-LAB Home
T-LAB PLUS 2019 - ON-LINE HELP Prev Page Prev Page
T-LAB
Introduction
What T-LAB does and what it enables us to do
Requirements and Performances
Corpus Preparation
Corpus Preparation
Structural Criteria
Formal Criteria
File
Import a single file...
Prepare a Corpus (Corpus Builder)
Open an existing project
Settings
Automatic and Customized Settings
Dictionary Building
Co-occurrence Analysis
Word Associations
Co-Word Analysis and Concept Mapping
Comparison between Word pairs
Sequence and Network Analysis
Concordances
Thematic Analysis
Thematic Analysis of Elementary Contexts
Modeling of Emerging Themes
Thematic Document Classification
Dictionary-Based Classification
Key Contexts of Thematic Words
Comparative Analysis
Specificity Analysis
Correspondence Analysis
Multiple Correspondence Analysis
Cluster Analysis
Singular Value Decomposition
Lexical Tools
Text Screening / Disambiguations
Corpus Vocabulary
Stop-Word List
Multi-Word List
Word Segmentation
Other Tools
Variable Manager
Advanced Corpus Search
Contingency Tables
Editor
Glossary
Analysis Unit
Association Indexes
Chi-Square
Cluster Analysis
Coding
Context Unit
Corpus and Subsets
Correspondence Analysis
Data Table
Disambiguation
Dictionary
Elementary Context
Frequency Threshold
Graph Maker
Homograph
IDnumber
Isotopy
Key-Word (Key-Term)
Lemmatization
Lexical Unit
Lexie and Lexicalization
Markov Chain
MDS
Multiwords
N-grams
Naïve Bayes
Normalization
Occurrences and Co-occurrences
Poles of Factors
Primary Document
Profile
Specificity
Stop Word List
Test Value
Thematic Nucleus
TF-IDF
Variables and Categories
Words and Lemmas
Bibliography
www.tlab.it

Test Value


This is a statistical measure which T-LAB uses to characterise two kinds of relationships:

a) those concerning lexical units with variable categories, the occurrences of which are summarized in contingency tables;
b) those concerning any row and column of a contingency table with the factors detected by a correspondence analysis of such a table.

Depending on the relationships analysed, the formulas of the test value, taken from one of the bibliography volumes (Lebart L. Morineau A. Piron M. , 1995, pp. 181-184), are the following:

a)

where 'njk' indicates the occurrences within a cell, while 'nj' and 'nk' correspond respectively to the marginal total of a row and of a column.

b)

where "nj" and " j" indicate respectively the occurrences of the j-th object and its coordinate on the -th factorial axis.

The test value has two significant properties: a threshold value (1.96), corresponding to the statistical significance most commonly used (p. 0.05), and a sign (-/+).

This means that, after sorting the values in an ascending or descending order, it is possible to quickly single out the relevance of each analysed item.

T-LAB allows a Test Value table consultation in an interactive way.