PROCESSING OF DATA
The data, after collection, has to be
processed and analysed in accordance with the outline laid down for the purpose
at the time of developing the research plan. Technically speaking, processing
implies editing, coding and tabulation of collected data so that they are
amenable to analysis.
The term analysis refers to the computation of
certain measures along with searching for patterns of relationship that exist
among data-groups. Thus, in the process of analysis, relationships or
differences supporting or conflicting with original or new hypothesis should be
subjected to statistical tests of significance to determine with what validity
data can be said to indicate any conclusions.
Importance of Data
Processing
Checks the data for Accuracy
Provides better understanding - it transforms
the data to information by classification, sorting, combination and reporting.
Puts into suitable form.
Helps in decision making
Makes data transferable
Readability and take less time.
PROCESSING OF DATA
Processing Operations or Data processing
refers to the process of converting data from one format to another. It
transforms plain data into valuable information and information into data. Data
processing services take the raw data and process it accordingly to produce
sensible information. The various applications of data processing include
converting raw data into useful information that can be used further for
business processes. Processing operations of data involves the following steps:
1) Editing,
2) Coding,
3) Classification, and
4) Tabulation.
EDITING OF DATA
By editing one tries to eliminate the errors
or points of confusion, there is no missing values, entries are readable, accurate
and uniform.
Stages in Editing
1. Field editing
2. Central Office
Editing
CODING OF DATA
In research methodology, coding is a stage in
data processing where qualitative data is labeled with descriptive keywords or
phrases to help identify and categorize related content.
It involves assigning number or symbols to
answers so that responses can be grouped a limited number of classes or
categories.
This process facilitates the
organization and analysis of qualitative data, enabling researchers to extract
themes, patterns, and relationships.
One purpose of coding is to transform the data
into a form suitable for computer-aided analysis
Role of Coding in
Research Process
Data Reduction: When researchers collect vast
amounts of data, coding helps condense and summarize it. This reduction makes
it feasible to analyze large datasets effectively.
Data Organization: Coding provides a
systematic way to categorize and group similar pieces of information together,
making it easier to manage and analyze the data.
Pattern Recognition: Coding allows researchers
to identify patterns, trends, and relationships within the data that might not
be immediately apparent when working with raw data.
Interpretation and Analysis: Coded data serves
as the foundation for statistical analysis and hypothesis testing. Researchers
can run statistical tests on coded data to draw meaningful conclusions.
Comparative Analysis: By coding data
consistently, researchers can compare and contrast information across different
cases or groups, aiding in the generation of insights and theories.
Examples of Data
Coding in Research
1. Qualitative Research
In qualitative research, data coding is often
used to categorize and analyze textual or narrative data. For instance, imagine
a study on customer feedback about a new product.
Researchers could code customer comments into
categories such as “product quality,” “customer service,” “pricing,” and
“delivery.” Each comment would be assigned one or more of these codes based on
the main topic it addresses.
2. Survey Research
In survey research, coding can involve
assigning numerical values to responses on a Likert scale. For example, in a
survey about job satisfaction, the responses “strongly agree” might be coded as
5, “agree” as 4, “neutral” as 3, “disagree” as 2, and “strongly disagree” as 1.
These codes enable quantitative analysis of survey data.
3. Content Analysis
Content analysis often involves coding textual
or visual content, such as news articles or social media posts, into predefined
categories.
For instance, in a content analysis of news
articles about climate change, researchers could code articles as “supportive
of climate action,” “neutral,” or “skeptical of climate change.” This coding
allows researchers to assess the prevalence of different perspectives in the
media.
4. Historical Research
Even in historical research, data coding can be useful. Historians might code historical documents based on themes, time periods, or key events. This enables them to identify patterns and trends across historical records and gain new insights into the past.
CLASSIFICATION
Classification is the process of arranging data in groups or classes based on common characteristics.
(a) Classification according to attributes also called statistics of attributes.
It can be descriptive such as: literary,
honesty etc. They are qualitative - Only their presence or absence can be
noticed.
It can be numeric such as-weight, height, income etc.
Classification can be simple classification where one attribute is considered and the the universe is divided into two, one processing the attribute and the other without.
Classification can be manifold classification where two or more attributes are considered simultaneously.
(b) Classification according to class intervals also called-. statistics of variables. -
Entire data is divided into number of groups or class or Class-Intervals.
Each group of class interval-has upper limit as well as a lower limit known an Class Limitd
The difference between the two class limits is
called class magnitude.
* How many classes should be there?
5 10 15 classes
* what should be their magnitude?
Multiples of 2, 5, 10 are generally preferred.
Intervals may be expressed as under 500 or
1001& over
How to choose the class limits?
The midpoint of a Class-Interval and the
actual average of items of that class interval should he as close as possible.
Exclusive type class - interval
10-20. Read as 10 and under 20
20-30 Read as 20 and under 30.
Inclusive type class-interval
11-20
21-30 etc.
