Tabulation and Data Analysis
Uploaded by
ScribdTranslationsTabulation and Data Analysis
Uploaded by
ScribdTranslationsData processing and analysis refers to the method of data analysis and
the techniques of statistical treatment.
The processing of information allows for organizing, creating graphs in certain types
of research and interpret all the information based on the proposals
theoretical, support for the conducted study. With the analysis of the information, elaborations are made.
the conclusions, recommendations and it structures the entire body of the work, which gives it
meaning to the research.
Once the fieldwork is completed and the obtained data is edited, it is necessary
carry out the processing of the data, that is, they are prepared to be
analyzed, for this, two data processing techniques are used: coding
and the tabulation.
Procedure:
1. Data Requirements Specification
The data required for the analysis is based on a question or an experiment.
Based on the requirements, the necessary data is identified, from the population or
data collection on the specific variables or attributes of them.
2. Data collection
Data collection in the data analysis process involves the gathering of
information about the selected variables as data requirements.
3. Data processing
The data that is collected must be processed or organized for analysis. This
includes structuring the data as needed for analysis tools
pertinent.
For example, it may be necessary to place the data into rows and columns in a
table within a spreadsheet or in a statistical application. It is possible that
you have to create a data model.
4. Data cleaning
At this stage of the data analysis process, it is carried out because processed data and
organized may be incomplete, contain duplicates or contain errors.
Data cleaning is the process of preventing and correcting these errors. There are several
types of data cleaning that depend on the type of data.
5. Data analysis
This is the fundamental phase of the data analysis process, which we reach with the
processed, organized, and clean data.
Various data analysis techniques are available to understand, interpret and
draw conclusions based on the requirements.
Data visualization can also be used to examine data in format
graph and to obtain additional information about the messages within the data.
Statistical data models such as correlation and regression analysis are
can be used to identify the relationships between data variables.
These descriptive models of the data are useful for simplifying the analysis and
communicate results. The process may require additional data cleaning or a
additional data collection, therefore these activities are of an iterative nature.
6. Communication
Once the fundamental phases of the data analysis process are completed, the
results must be presented in a format as required by the users for
support their future decisions and actions.
User feedback could result in additional analysis.
Data analysts can choose data visualization techniques such as
tables and graphs, which help to communicate the message clearly and efficiently to the
users.
As previously expressed, in order to carry out the analysis process
It is necessary to encode the data. Therefore, Encoding: means assigning a
code, generally a number, to each possible answer to each question. The code
include an indication of the position in the column (field) and the record that it will occupy
data. Example: If the database includes the variable Gender, a number must be assigned to it.
women and another for men. If it is about quantitative variables, it is necessary to define the
number of decimals that will be recorded.
On the other hand, a practical way to enter the data is to use a spreadsheet like
EXCEL, which allows for easy assignment of specific columns for questions and
concrete answers. Each row contains the data of a respondent.
Procedure:
It is necessary to transcribe the information into an analytical memo (recording technique of
information), which encompasses the most important categories, questions and
responses, keywords, and some subcategories that arise from the results of the
relationship of the identified key words.
This table allows for the initial coding, which enables the
identification of central ideas that are presented in the data and that are represented
as concepts.
2. Axial coding: Key words are taken up to make a relational reading.
from the data, that is, each keyword is taken and based on it, the
identification of matches and relationships in the participants' responses.
3. Selective coding: In this step, a relationship is established again, but between the
key words that emerged from axial coding, seeking to regroup
in subcategories. Therefore, selective coding consists of continuing the analysis
until the concepts are validated; achieving integration into category
emergents.
After creating the data matrix, the tabulation of the same is proceeded.
According to Rojas Soriano, R: 'Tabulation is the process by which data
collected are organized and concentrated, based on certain ideas or hypotheses, in
tables or charts for their statistical treatment.
In short, data tabulation refers to the coding and transcription of
the data obtained in the market research.
Procedure:
Detail the identified variables.
2. Determine the variables that need to be analyzed individually.
3. Determine the variables that should be crossed according to the objectives.
4. Outline the table to determine the possibility of variable crossing,
according to the number of these that must be related and the scales of categories of
analysis.
5. Make the list of the paintings that must be presented.
Cross tabulations: Statistical technique that describes two or more variables in a way
simultaneous, and produces tables that reflect the joint distribution of two or more
variables with a limited number of categories or distinctive values.
When carrying out a cross-tabulation analysis, it is helpful to follow these steps.
1. Test the null hypothesis that there is no association between the
variables through the chi-square statistic. If the hypothesis is not rejected
none, so there is no relationship at all.
2. If H0 is rejected, then the strength of the association must be determined.
using an appropriate statistic (phi coefficient, contingency coefficient,
In Cramer's V, lambda coefficient or another statistic).
3. If H0 is rejected, the pattern of the relationship must be interpreted by calculating the
percentages in the direction of the independent variable, by the variable
dependent.
4. If the variables are treated as ordinal rather than nominal, tau is used.
b, tau c or gamma as test statistic. If H0 is rejected, then it
determine the strength of the association using the magnitude and direction of the
relationship with the sign of the test statistic.
5. Translate the results of the hypothesis test, the strength of the association and the
pattern of the association, implications and recommendations for management
when convenient.
