|
Data Processing
Data processing and analysis is performed
on the "Basic
Module"
and it falls into two stages: preliminary and mathematical.
1. Preliminary Processing
On
this stage a researcher has an opportunity to prepare "data sets"
that represent the calculated semantic coordinates (see below), proceeding
from the tasks relevant to researcher.
Conditions for set formation:
Data filtration allows
the researcher to classify expert-submitted data in terms of a definite
age group, gender, category, and calendar date of data input. An option to restrict
the subject content and scales out of the ones available in a project
(with the purpose of the most flexible and detailed problem study). Setting a number of analyzed
subject states that characterize independent sense variants of subject
perception within the studied mentalities. Selection of bilateral
significance level for the criterion for rejecting the data from marginal
experts of the studied mentality.
Set preparation
consists of the following stages:
Creation
of a data set starts with the calculation of a subject's semantic coordinates
for every separate expert. Rejection
of semantic descriptions and subject priorities that do not comply with
the specified significance level (rejection of marginal experts who gave
estimates discordant with those of the group). Calculation
of semantic coordinates, rigidities of the selected number of states as
well as probabilities of realizing these states for each subject after
all the first-stage data has been collected. Entering
the result set into the database for further use in mathematical analysis
with set parameters indicated (conditions of noise filtration, expert
categories, scales, subject list, etc.)
A researcher has an option to create
an arbitrary number of sets, which gives him the opportunity to concentrate
on the information (mathematical) data analysis, without returning to
the initial data every time. The proposed data storage structure makes
it possible to save initial data intact, which opens the way for its extension
due to the experts actually participating in the project or potentially
involved.
2. Mathematical Analysis
Using
this mode, a researcher can calculate the following characteristics and
parameters of the selected mentality:
Relative
and absolute semantic coordinates of the states (subjects) and their probabilities. Rigidity
of properties and states. Factor
weights of the properties/characteristics (factor significance) and factor
semantic representation of subject states (factor weights of research
subjects). After
"varimax"-rotation
of the factors one can get categorial weights of the properties and categorial
semantic representation of subject states. Semantic
representation of a motive in factor and categorial representation. Weight
of states in research subject priorities. Weights
of factors and categories (factor
and category contribution) in research subject priorities. Degree
of subject's proximity to the motivational vector.
|
