31/01/2015 CONTENT ANALYSIS

CONTENT ANALYSIS

Roque Moraes

MORAES, Roque

Summary

This article aims to present a methodology for analyzing qualitative data,

generally referred to as content analysis. Based on a set of principles characterizing
this methodology, the raw material it works with, as well as the different objectives it can achieve
focalize, a description of a set of steps is presented according to which it can be conceived
It is applied. In this sense, categorization, description, and interpretation stand out as essential steps.
this analysis methodology. The work concludes by discussing different possibilities of application
of this methodology.

Abstract

This paper presents a methodology for qualitative data analysis, usually called content analysis.
Beginning with a characterization of this methodology, the type of data it works with, as well as the
different objectives it may be focused on, a description of steps according to which this is presented
analysis can be conducted. Categorization, description and interpretation are emphasized as essential
steps in this methodology of analysis. The discussion is finished presenting different possibilities of
utilization of this methodology.

Introduction

Content analysis has its origins in the late last century. Its characteristics and different
approaches, however, have been developed, especially over the last fifty years.

Even though it was a phase of great productivity when it was guided by the paradigm
positivist, greatly valuing objectivity and quantification, this methodology of analysis
data is reaching new and more challenging possibilities as it integrates more and more
but in the qualitative exploration of messages and information. In this sense, even if eventually
not under the designation of content analysis, it insinuates itself in works of a dialectical nature,
phenomenological and ethnographic, among others.

We seek to explore a set of information about nature and the

characteristics of content analysis. From this, we focus on some of the procedures that we
uses in the examination of the materials it deals with. We conclude our discussion by addressing some of the
strategies and possibilities that present themselves to the researcher interested in using it.

Our goal is to propose a discussion that is both comprehensive and condensed about nature,
methodology and possibilities of content analysis within educational research.

Introduction

Content analysis originated in the late last century. Its characteristics and different
approaches, however, have been developed, especially over the last fifty years.

Even though it was a phase of great productivity when it was guided by the paradigm
positivist, highly valuing objectivity and quantification, this analysis methodology of
data is reaching new and more challenging possibilities as it integrates increasingly
but in the qualitative exploration of messages and information. In this sense, even if eventually
not with the designation of content analysis, it insinuates itself in works of a dialectical nature,
The provided text is a URL and does not contain translatable content. 1/13
31/01/2015 CONTENT ANALYSIS

phenomenological and ethnographic, among others.

We seek to explore throughout this text a set of information about nature and the
characteristics of content analysis. From this, we focus on some of the procedures that are
uses in the examination of the materials it deals with. We conclude our discussion by addressing some of the
strategies and possibilities that arise for the researcher interested in using it.

Our goal is to propose a discussion that is both comprehensive and concise about nature,
methodology and possibilities of content analysis within education research.

Fundamentals

Content analysis is a research methodology used to describe and interpret the

content of all classes of documents and texts. This analysis, leading to systematic descriptions,
qualitative or quantitative, helps reinterpret the messages and achieve an understanding of their
meanings on a level that goes beyond a common reading.

This research methodology is part of a theoretical and practical search, with a special meaning.
in the field of social investigations. It constitutes much more than a simple analysis technique
of data, representing a methodological approach with its own characteristics and possibilities.
In its evolution, content analysis has oscillated between the rigor of the supposed objectivity of
numbers and the ever-questioned fertility of subjectivity. However, over time, they have
qualitative approaches have increasingly been valued, especially using induction and
intuition as strategies to achieve deeper levels of understanding of the phenomena that
aims to investigate.

In any of its approaches, it provides complementary information to the critical reader of a

message, be it linguist, psychologist, sociologist, educator, literary critic, historian or another.

As a research method, content analysis encompasses special procedures for the

processing of scientific data. It is a tool, a practical guide for action, always renewed
due to the increasingly diverse problems it proposes to investigate. It can be considered
it as a single instrument, but marked by a great variety of shapes and adaptable to a
very broad field of application, namely communication.
According to OLABUENAGA and ISPIZÚA (1989), content analysis is a technique for reading and
interpret the content of all classes of documents, which when properly analyzed open up to us
opens up knowledge of aspects and phenomena of social life in a way that was otherwise inaccessible.

The raw material for content analysis can consist of any material derived from
verbal or non-verbal communication, such as letters, posters, newspapers, magazines, reports, books, accounts
autobiographical, albums, recordings, interviews, personal diaries, films, photographs, videos, etc.
However, the data from these diverse sources comes to the investigator in raw form,
needing to be processed in order to facilitate the work of understanding,
interpretation and inference that content analysis aspires to.

Content analysis, in its qualitative aspect, is based on a series of assumptions, which, in

The examination of a text serves as support to capture its symbolic meaning. This meaning is not always
The manifesto and its meaning is not unique. It can be focused on from different perspectives.
Therefore, a text contains many meanings and, as OLABUENAGA and ISPIZÚA state
(1989, p.185):
(a) the meaning that the author intends to express may coincide with the meaning perceived by the reader.

(b) the meaning of the text may vary according to each reader;

(c) the same author may send a message, and different readers may capture it with
different senses;

The provided text appears to be a URL and does not require translation. 2/13
31/01/2015 CONTENT ANALYSIS

A text can express a meaning of which the author himself is not aware.

In addition, it is important to emphasize that it will always be possible to investigate the texts within multiple
perspectives, as expressed by KRIPPENDORF, 1990, p.30:
In any written message, letters, words, and sentences can be computed simultaneously;
categorize the sentences, describe the logical structure of the expressions, verify the associations, denotations,
connotations and may also formulate psychiatric, sociological or political interpretations.

It should still be remembered that, although the consensus or intersubjective agreement on what it means a
simplify the content analysis, this coincidence of meanings is not indispensable.
The values and natural language of the interviewee and the researcher, as well as the cultural language and
their meanings exert an influence on the data from which the researcher cannot escape.

In a certain way, content analysis is a personal interpretation by the researcher.

regarding the perception you have of the data. A neutral reading is not possible. Every reading is constituted
in an interpretation.

This issue of multiple meanings of a message and the multiple possibilities of analysis
what enables is very closely related to the context in which the communication takes place.

Even though in its original proposal content analysis was more specifically concerned with the
The meaning of messages for recipients, in their evolution, has taken on increasing importance.
greater investigations focusing on both the process and the product, considering both the emitter
like the receptor.

Throughout this evolution, the understanding of context increasingly proves to be indispensable.

to understand the text. The message of communication is symbolic. To understand the meanings of a
Text, therefore, it is necessary to take context into account. It is necessary to consider, in addition to the content
explicit, the author, the recipient, and the forms of encoding and transmission of the message.

The context within which the data is analyzed must be made explicit in any content analysis.
Although the data is expressed directly in the text, the context needs to be reconstructed by the
researcher. This establishes certain limits. It is not possible to include, in this reconstruction, all the
conditions that coexist, precede or succeed the message, in time and space. They do not exist.
logical limits to delimit the context of the analysis. This will depend on the researcher, the discipline and
Two objectives proposed for the investigation, in addition to the nature of the materials under analysis.

Considering the mentioned aspects, a research using content analysis needs

to be based on a clear explanation of its objectives.

Depending on the research approach used, the definition of objectives can take two paths.
different. In a quantitative, deductive approach for hypothesis verification, the objectives are
defined in advance in a quite precise manner. They constitute an essential part of the initial planning
which precedes and guides the later phases of the research, especially the definition of the data and the
specific analysis procedures. In a qualitative, constructive or heuristic approach, this
construction, at least in part, can occur throughout the process. In this approach, just as the
categories may emerge throughout the study, as well as the more specific guidance of the work,
the objectives in their most precise sense may begin to take shape as the research progresses
it advances. However, generally speaking, it is possible to affirm that upon the completion of a research
It is important to be able to clearly articulate the objectives of the work done. When one uses the
content analysis, a clear statement of objectives helps to delimit the data effectively
significant for a certain research.

Naturally, there would be many ways to categorize possible research objectives.

using content analysis. However, historically these have been defined in six categories,
taking into account the intrinsic aspects of the raw material of this analysis, of the context to which the
Research refers to and the intended inferences. This classification is based on a definition
original by Laswell, in which he characterizes communication based on six questions: 1) Who speaks? 2)
To say what?
Unable to access the provided URL for translation. 3/13
31/01/2015 CONTENT ANALYSIS

Using this definition, we can categorize the objectives of content analysis according to the
guidance that you take in relation to these six issues.

A Content Analysis oriented to 'who speaks?' aims to investigate who is issuing the message.
study will naturally be carried out from the message, from which an attempt will be made to determine
characteristics of those who speak or write, whether regarding their personality, verbal behavior,
values, semantic universe, psychological characteristics or others. In this case, in a way, it advances-
if the hypothesis that the message expresses and represents the sender. In light of this objective, it is necessary to
inferences from the text to the sender of the message.

When a research using content analysis addresses the question to say what? the study does
focuses on the characteristics of the message itself, its informational value, the words,
arguments and ideas expressed in it. This is what constitutes a thematic analysis.

The objectives of content analysis research can also be guided by 'to whom?' it is directed.
the message. In this case, the investigation focuses on the receiver, trying to infer the characteristics of this one,
based on what you read or hear. Indicators and characteristics of the message lead to inferences about
who receives it.
When the objectives of the analysis direct to the 'how?', the researcher will focus on the way in which
communication takes place, its codes, its style, the structure of the language and other characteristics
of the medium through which the message is transmitted.

By directing your study to the 'for what purpose?' the researcher will question about the
objectives of a given communication, whether explicit or implicit. Analyzing the purpose is
again orient oneself towards the issuer, but in the sense of capturing the purposes with which one issues a
determined messages, whether they are manifest or hidden.
By focusing on the 'with what results?', the researcher seeks to identify and describe the results.
the effects of communication. The objectives do not necessarily coincide with the results
effectively achieved and, thus, the research can also explore the issue of congruence between ends
the results.

The definition of objectives for a content analysis based on this scheme or another does not imply
by adhering to one of these categories. Research may be directed at the same time towards two or
but of these issues. Similarly, the methods and techniques of analysis may vary depending on
two proposed objectives. The perception of the content and especially the inferences reached
depend on the proposed objectives.

An understanding of the fundamentals of content analysis is certainly important for the analyst.
to get the most out of this methodology. Understand its history, understand the types of materials
that enables analysis, while being aware of the multiple interpretations that a
The message always enables, leading to the understanding of a multiplicity of objectives that a
content analysis can reach, helping to better explore the possibilities of this methodology of
analysis.

We will now discuss more specific aspects of the methodology that underpins the analysis of
content.

Method

Although different authors propose various descriptions of the analysis process

content, in the present text we conceive it as consisting of five stages:

1 - Preparation of information;

2 - Unitarization or transformation of content into units;

3 - Categorization or classification of units into categories;

The provided text is a URL, which cannot be translated. 4/13

31/01/2015 CONTENT ANALYSIS

4 - Description;

5 - Interpretation.

When we discuss these stages, we are mainly focusing on a qualitative analysis, even
understanding that they can also be applied in quantitative studies.

Throughout this sequence we are only referring to the analysis itself. These five
stages must necessarily be preceded by the normal definitions that accompany a project
of research, namely to make a problem explicit, to clearly establish the objectives of the research and the
from this, gather the data forecasted by the project.

We will examine each of these steps next.

1 - Preparation

Once in possession of the information to be analyzed, it is necessary first to submit it to a

preparation process. This consists of:

1.1-Identify the different samples of information to be analyzed. For this, it is recommended that a
reading all the materials and making a first decision on which of them are actually relevant
in accordance with the research objectives. The documents thus included in the sample must be
representative and relevant to the objectives of the analysis. They should also cover the field to be
investigated comprehensively.

1.2-Start the process of coding the materials by establishing a code that allows
quickly identify each element of the sample of testimonies or documents to be analyzed.
This code may consist of numbers or letters that from this moment will guide the
researcher to return to a specific document whenever desired.

Even though the documents to be examined through content analysis have already been
they need to be prepared and transformed to constitute the information to be
submitted to content analysis. The data is not entirely data, but needs to be
properly prepared for such. In this, the research objectives take on a central role.

2 - Unitarization

Once properly prepared, the data will undergo the process of 'unitarization'.
consists of the following:

2.1 - Carefully reread the materials in order to define the unit of analysis. Also
referred to as 'unit of record' or 'unit of meaning', the unit of analysis is the element
unit of content to be subsequently submitted for classification. All categorization or
classification, it is necessary to define the element or individual unit to be classified. In the analysis of
we call this element the unit of analysis.

The nature of the units of analysis needs to be defined by the researcher. The units can be
both the words, phrases, themes or even the documents in their integral form. In this way for the
definition of the units of analysis constituting a set of raw data can maintain the
documents or messages in their entirety or they can be divided into smaller units. The
The decision about what the unit will be depends on the nature of the problem, the research objectives and
of the type of materials to be analyzed.

2.2 - Review all materials and identify the units of analysis within them. By doing so
each unit is coded, with additional codes established, associated with the coding system
already prepared previously. Upon completing this process, one usually has the different messages
divided into smaller elements, each identified by a code that specifies the unit
from the sample from which it originates and within it the sequential order in which it appears. Thus, for example, if
a sample document received the code '1', the different units of analysis of this document
they can receive the codes '1.1', '1.2', '1.3' and so on.

The provided text is a URL and does not contain translatable text. 5/13
31/01/2015 CONTENT ANALYSIS

2.3-Isolate each of the units of analysis.

Each unit of analysis needs to be isolated in order to be submitted for classification.

It is recommended that this be done by rewriting each of them on a card in order to make them
individualized and isolated. This process of isolating the units of analysis subsequently requires that
these should be rewritten or reworked so that they can be understood outside of the context
original in which they were located.

In the process of transforming raw data into units of analysis, it is important to take into account that
these must represent sets of information that have a complete meaning in themselves
the same. They should be able to be interpreted without the assistance of any additional information. This is
important, since these units in the later phases of the analysis will be treated outside the context of
original message, integrating into new sets of information and then, they should be able to
to be understood and interpreted while maintaining the original meaning.

It is important to emphasize that in this process of fragmenting a text, part of it is necessarily lost.
of the analyzed material. The reading made will always represent a perspective of
researcher. However, as one becomes aware that there is no objective reading
and completion of a text, this loss of information can be justified by the deepening in
understanding that the analysis enables.

2.4-Define the context units

Although it is desirable and important to seek to define the units of analysis in such a way that they have a
complete meaning in themselves, it is also a practice of content analysis to define, along with
with these units, another type of content unit, the context unit. It is a unit of
broader general mode than that of analysis, which serves as a reference for it, establishing contextual limits
to interpret it. Each unit of context generally contains several units of record.

The justification for the proposal of the context units is based on the already expressed conviction
that, even though a message can be divided into independent units of meaning, the
units of analysis, meanings will always be lost in this process. That is why it is important to be able to
periodically return to the context from which each unit of analysis comes, in order to explore
in a more complete way all its meaning.

Once all the units of analysis have been identified and coded, the content analyst will be ready
to get involved with categorization. In fact, continuously, especially if the quantity of
the materials to be investigated are numerous, it is recommended to carry out the unitarization work initially
only with part of the material. Then a first effort of categorization is made, returning
after the unitarization to complete the work. This is especially true when the categories
are defined based on the material under examination and when the very concept of unit of analysis is
built from the investigated content.

3 Categorization

Categorization is a procedure for grouping data considering the common part that exists among
They are classified by similarity or analogy, according to previously established criteria or
defined in the process. These criteria can be semantic, originating thematic categories. They can
syntactic beings defining categories based on verbs, adjectives, nouns, etc. The categories
can still be constituted based on lexical criteria, with an emphasis on words and their meanings or
they can be founded on expressive criteria focusing on language problems. Each set
Categories, however, must be based on only one of these criteria.
Throughout this text, we will give special emphasis to the thematic categories.

As OLABUENAGA and ISPIZÚA (1989) state, the process of categorization must be understood
in its essence as a process of data reduction. The categories represent the result of a
synthesis effort of a communication, highlighting its most important aspects in this process.

Categorization is, therefore, an operation of classifying the elements of a message.

The provided text is a URL and cannot be translated. 6/13
31/01/2015 CONTENT ANALYSIS

following certain criteria. It facilitates the analysis of information, but it must be based on a
precise definition of the problem, objectives, and elements used in content analysis.
It is also necessary to understand that the analysis of the material occurs in a cyclical and circular manner, and
not in a sequential and linear way. The data does not speak for itself. It is necessary to extract meaning from it.
This is generally not achieved in a single effort. The periodic return to the data, the refinement
progressive of the categories, within the search for meanings that are increasingly well clarified,
they constitute a process never entirely complete, in which each cycle can reach new
layers of understanding.

But what characteristics should the categories have? What criteria should the categorization follow?
content analysis?

Categorization is undoubtedly one of the most creative stages of content analysis. However, be
with predefined categories, whether with a categorization based on data, the establishment
Categories must comply with a set of criteria. The categories must be valid.
exhaustive and homogeneous. The classification of any element of the content must be mutual.
exclusive. Finally, a classification must be consistent. Even admitting differences in
application and interpretation of these criteria, it is important to discuss and understand them. The eventual not
the service to any of them in a survey must be properly justified.
The amplitude and precision of the categories are directly related to the number of categories: in general,
the more subdivided the data and the greater the number of categories, the greater the accuracy of the
classification. However, it is necessary to take into account that a large number of categories can
introduce comprehension difficulties. The basic objective of content analysis is to produce a
reduction of the data from a communication, which will generally require a reduced number of categories.

The main criteria for the constitution of categories in content analysis are discussed below.

First of all, the categories need to be valid, relevant, or appropriate.

An initial and basic characteristic of any set of categories must be its validity.
To say that a categorization must be valid means to say that it must be appropriate or relevant. This
adaptation refers to the objectives of the analysis, the nature of the material being analyzed, and the
questions that one intends to answer through the research. Validity or relevance requires that all the
created categories are meaningful and useful in terms of the proposed work, its problems, its
objectives and their theoretical basis. In addition, all significant aspects of the content
investigated and the objectives and problems of the research must be represented in the categories. Among
the two extremes, creating new categories and creating useful and meaningful categories, it is necessary to achieve a
balance in which the number of categories is kept to the minimum necessary. The researcher must be
frugal in this sense.

When categories are defined a priori, validity or relevance can be constructed from
a theoretical foundation. In the case where categories emerge from the data, the arguments of validity
are built gradually. A valid categorization must be meaningful in relation to the
contents of the materials being analyzed, constituting an appropriate reproduction and
relevant to this content.

A second criterion to be met in categorization is exhaustiveness or inclusivity.

If we can affirm that the basic rule of categorization is that of validity, another rule follows from it,
to the exhaustiveness. Saying that a set of categories must be exhaustive means that it must
enable the categorization of all significant content defined according to the objectives of
analysis. Thus, each set of categories must be exhaustive in the sense of allowing the inclusion of
All units of analysis. There should be no significant data that cannot be classified.
It is never too much to remind that the rule of exhaustiveness needs to be applied to the contents effectively.
significant of the study. The objectives of the analysis define the dataset that effectively
they should be categorized. Once this decision is made, the categories should be exhaustive, that is
to have the possibility to fit all the content.

URL provided does not contain translatable text. 7/13

31/01/2015 CONTENT ANALYSIS

The categories must also meet the criterion of homogeneity.

In addition to being valid and sufficiently comprehensive to allow for the inclusion of all
meaningful data, the categories also need to be homogeneous. Their organization must be
based on a single principle or classification criterion. To say that a set of
categories being homogeneous means being able to affirm that the whole set is structured in a single
dimension of analysis. From a quantitative perspective, it is to say that it must be based on a single variable.

If there is more than one level of analysis, the criterion of homogeneity must be present at all levels.
levels. Furthermore, it is important that this homogeneity is not only guaranteed in content but
similarly at the level of abstraction.

In addition to the previous criteria, the categories must also meet the exclusivity criterion or
mutual exclusion.

Ensured the exhaustiveness and homogeneity of its categories, the content analyst needs to
ensure that each element can be classified into only one category. It is the criterion of
exclusivity or mutual exclusion. The same data cannot be included in more than one category,
That is, each element or unit of content cannot be part of more than one division.

For a set of categories to meet the exclusivity criterion, it is important that it be

built a set of very precise and clear classification rules.

Finally, the categories in content analysis must meet the criterion of objectivity.
consistency or fidelity.

The criterion of objectivity or consistency is closely related to the criterion of

exclusivity. When a set of categories is objective, the classification rules are
explicitly stated with sufficient clarity so that they can be consistently applied throughout
all the analysis. This means that there should be no doubt regarding the categories in which each
Content unit should be integrated.

When a set of categories meets the criterion of objectivity, the classification will not be affected.
through the subjectivity of the coders. Under these conditions, different researchers will have to reach
similar results when categorizing the same content units, based on the same
classification rules.

The issue of objectivity has been increasingly questioned in qualitative research. In this type
In research, the researcher often believes in multiple realities and each one can capture.
different dimensions from the same text or message. This, however, will be reflected
possibly more in the constitution of different sets of categories than in the classification of
content, once the set of categories is established. Therefore, even in research
essentially qualitative, the criterion of objectivity or consistency of the categories and of the
classification seems to remain significant and important to be considered.

To conclude this quick discussion on categorization, its meaning and the criteria for its establishment
categories it is important to remember that data can be grouped within various levels of
categorization. When this occurs, the criteria must be applied at each of the levels. The
categories resulting from the first classification effort, generally more numerous, homogeneous
and you need, they can be called initial categories. Those that come from regrouping.
progressive, with a weaker homogeneity, in smaller numbers and broader, may be
referred to as intermediate and final categories.
It is still important to emphasize that the construction of validity and the other characteristics of the categories of
A content analysis, especially in a qualitative analysis, occurs throughout the entire process.
A priori defined categories must already meet the classification criteria beforehand, that is, before.
to proceed with the actual classification of the content. Categories constructed from the very own
materials require that compliance with classification criteria occurs throughout the process of
analysis. The arguments of validity, exhaustiveness, homogeneity, exclusivity, and objectivity
they need to be built throughout the analysis.

Unable to access external URLs for translation. 8/13

31 January 2015 CONTENT ANALYSIS

4 - Description

The fourth stage of the content analysis process is description. Once the categories are defined and
Once the constituent material of each of them is identified, it is necessary to communicate the result of this work.
The description is the first moment of this communication.

When it comes to research using a quantitative approach, this description will involve the
organization of tables and charts, presenting not only the categories constructed in the work, but also
also computing frequencies and percentages related to them. There may be different types
of the tables, according to the levels of categorization used.

When it comes to research in a qualitative approach, the description will generally be different.
order. For each of the categories, a summary text will be produced that expresses the set
of meanings present in the various units of analysis included in each of them. Generally, it is
It is advisable to make intensive use of 'direct quotes' from the original data.
This will not yet be the interpretive moment, although there may be increasingly more descriptions.
comprehensive, depending on the levels of categorization. Generally speaking, the organization of this
description will be determined by the category system built throughout the analysis.
The moment of description is undoubtedly of extreme importance in content analysis. It is the moment
to express the meanings captured and intuited in the analyzed messages. It is pointless to invest too much
time and effort in the constitution of a significant and valid set of categories, if at the moment
to present the results if the same care is not taken. It will generally be through the text.
produced as a result of the analysis that one can understand the validity of the research and its
results.

The chapter of a research report in which the descriptions of the analysis results are presented
the content is certainly one of the most important chapters of this report. However, it is not
sufficient. It is required to reach the interpretation.

5 - Interpretation
A good content analysis should not be limited to description. It is important to seek to go beyond,
achieve a deeper understanding of the content of the messages through inference and
interpretation.

The term infer refers more specifically to quantitative research. The inferential hypothesis test.
establishes the limits within which the findings of a study, usually based on a sample, are
generalizable to the population from which the sample comes. Infer from the sample to the
the population is, therefore, this extension of the conclusions of a smaller group to a larger population
ample.
The term interpretation is more associated with qualitative research, although it is not absent in the approach.
quantitative. It is linked to the movement of seeking understanding. Every reading of a text constitutes
in an interpretation. However, the content analyst exercises this effort with greater depth.
of interpretation and it does so not only about content manifested by the authors, but also about the
latent, whether they are consciously or unconsciously hidden by the authors.

In the interpretive movement, we can highlight two aspects. One of them relates to studies with
a theoretical foundation clearly stated a priori. In these studies, the interpretation is made
through an exploration of the meanings expressed in the analysis categories in a contrastation
with this foundation.

On the other hand, the theory is built on the data and the categories of analysis. The theory
emerges from the information and from the categories. In this case, the very construction of the theory is a
interpretation. Theorization, interpretation, and understanding constitute a circular movement in which the
With each iteration of the cycle, the aim is to achieve greater depth in the analysis.

In any case, whether from a defined theoretical foundation a priori, or from the production
From theory based on the materials under analysis, interpretation is an essential step in all
The provided text is a URL, which cannot be translated. September 13
31/01/2015 CONTENT ANALYSIS

the content analysis, especially those of a qualitative nature.

Approaches

Content analysis allows different ways to conduct the process. They are at the same time
options in terms of opening possibilities that this methodology offers, establishing itself in the same
movement limits as a consequence of these options.
One of these possibilities refers to the choice that the researcher can make regarding the type of content.
which aims to examine. It will limit itself to the manifest content or will also seek to explore the
latent content? This option also relates to the choice for an objective exploration as opposed to
to a subjective analysis.

Content analysis at the manifest level is limited to what is said, without seeking the meanings.
hidden. At the latent level, the researcher seeks to capture implicit meanings. Following that, the analysis of
content is part of the information manifested in the text in order to then address the intention that the author wanted
expressing, sometimes reaching the point of capturing something that even the author was not fully aware of.

The manifest and latent levels are related to the emphases on objectivity or subjectivity, between
to which content analysis oscillates. The manifest level corresponds to a representational reading,
in which the direct inference of what the author meant is sought. But the search for an understanding
but deeper cannot ignore the latent content of messages, revealed before by the unspoken that
by the express. This corresponds to a reading that captures unconscious motivations in the lines.
indiscernible, revealed by discontinuities and contradictions.

In relation to this, one can question: How to conduct a valid data analysis?
subjective? According to LINCOLN and GUBA (1982), the dimension of objectivity-subjectivity does not
questions the objectivity or subjectivity of the researcher. It corresponds to the way of categorization,
because conceptual categories can be established a priori to be applied to the text, the
objective approach, or they can constitute an inductive process, reconstructing the categories
used by individuals to express their own experiences and worldview, the approach
subjective.
The emphasis on subjectivity is not incompatible with scientific rigor. This does not exclude or replace
latent senses and unquantifiable intuitions. Content analysis, in a qualitative approach,
surpasses the manifest level, articulating the text with the psychosocial and cultural context.

Naturally, the choice of one or the other of these alternatives also corresponds to the assumption of
certain epistemological and research assumptions, whether they are made explicit or not.

The previous discussion, leading us through the manifest and latent levels of the contents of
messages, necessarily related to the issue of objectivity and subjectivity, confronts us
to what we could call two basic approaches to content analysis. One of them can be
identified, using a characterization proposed by Getz and LeCompte (cited in Lincoln & Guba,
1982) comeductive, verificatory, enumerative, and objective. The other, at its extreme, is characterized by
for being inductive, generative, constructive, and subjective. This bipolarization can be interpreted as
related to the preferential use of deductive versus inductive reasoning. The former seeks
explanations and probabilistic generalizations. The second aims at understanding phenomena.
investigated. The deductive approach starts from a theory, while the inductive approach aims to reach a theory.

Depending on the approach on which the researcher bases their research, there will be
consequences not only on analysis procedures but also on other aspects
important in research.
The deductive-verificatory-enumerative-objective approach starts from theories and hypotheses proposing a
testing or verification of them within the canons of traditional research. The hypotheses help
to direct the process, defining together with the theory the nature of the data and its organization.
For this reason, in this approach the categories are provided or established a priori, either from the
theory, whether of objectives or research questions. In any case, in this approach the
categories need to be justified based on a solid theoretical foundation. The adoption of this
URL provided is not translatable text. October 13
31/01/2015 CONTENT ANALYSIS

approach, in general, implies the use of quantification as the preferred treatment of data,
culminating in hypothesis tests that may lead to justified inferences within standards
quantitative research statisticians. In this approach, objectivity is also highly emphasized.
considered, even acknowledging that this may result in the loss of precious information that
Perhaps the messages load, but do not fit into the categories defined a priori.

Undoubtedly, the deductive-verificatory approach allows for achieving levels of accuracy, rigor and
more acceptable systematization within the perspective of traditional research. However, this option does not stray
to a whole set of criticisms that have been raised about this type of research and its limitations.

The inductive-constructive approach starts from the data, building from them the
categories and from these the theory. It is therefore essentially inductive. Its purpose is not
to generalize or test hypotheses, but to build an understanding of the investigated phenomena.

In this approach, the categories are built throughout the process of analysis. The categories are
resulting from a progressive and analog systematization process. The emergence of categories is
result of an effort, creativity, and insight on the part of the researcher, requiring a reinterpretation
exhaustive in defining what is essential based on the proposed objectives. The titles of the categories
only appear at the end of the analysis. For this approach, according to LINCOLN and GUBA (1985), two
Strategies have been suggested: the method of analytical induction and the method of constant comparison.
Both are based on induction, a process in which the rules of categorization are
developed throughout the analysis, and make intense use of the researcher’s tacit knowledge
as a basis for the constitution of the categories. Both the categories and the rules of
Categorization is continuously reviewed and refined throughout the entire analysis.

The inductive-constructive approach also pursues scientific rigor, albeit not strictly defined.
in the same way that the deductive-verifying approach does. The rigor and scientific rigor from the perspective
Inductive reasoning needs to be built throughout the entire process and cannot be guaranteed a priori.

To conclude the discussion on the two approaches to content analysis presented, deductive-
verificatory and inductive-constructive, it is important to emphasize and deepen the issue of the role and the
meaning of the theory in each of them. In the first, the theory precedes the analysis and serves as a foundation
for her. On Monday, the theory emerges from the analysis, that is, it results as one of its products. This
characterizes two extremes, allowing for research that addresses the question of the meaning of
theory and theoretical foundation of intermediate perspectives between these two extreme positions.

These are some considerations about specific aspects of content analysis methodology and
its approaches. A deeper understanding, both of the fundamentals and especially
this analysis methodology is only acquired through practice. As this understanding
advance, the researcher will also feel the need to position themselves more clearly on issues
epistemological. We understand that content analysis constitutes a versatile tool,
can be adapted to a wide range of research, especially in studies in the area
social. Each researcher is challenged to try to find the way of its use in specific areas
of your work.

Final considerations

The aim of this text was to discuss some elements of content analysis. It began
focusing on some aspects of its characterization and foundations. Its raw material and the
meaning of context in text analysis. A diverse set of was also demonstrated.
objectives that this analysis methodology allows to achieve.

Then they began to examine the methodological question. In this sense, coding was emphasized.
transformation of the text into units of analysis and their categorization. The criteria were also discussed
a good set of categories. The stages of description and interpretation were emphasized as essential
in content analysis.
Finally, the possibilities and variations of this analysis were discussed. Analyses based on were identified.
manifesto content and made a comparison with proposals for studying the latent content of the
messages. The possibilities of deductive, verificatory, and objective analyses were contrasted with
The provided text is not translatable content. November 13
January 31, 2015 CONTENT ANALYSIS

inductive, constructive, and subjective analyses. This discussion sought to visualize some of the
different possibilities that this analysis methodology offers to the researcher.
At the conclusion of this discussion, the aim is to have gathered arguments that show that the analysis of
content is not constituted by a simple technique, but can be considered as a
varied methodology and in continuous review.
In this sense, it is understood that content analysis allows for the fulfillment of numerous
needs of researchers involved in communication data analysis, especially those
focused on a qualitative approach.

Bibliography

BARDIN, L. Content analysis. Lisbon, Edições 70, 1977.

BARROS, A.J.P.; LEHFELD, N.A.S. Research project: methodological proposals. Petrópolis,
Voices, 1991.

DELGADO, J.M.; GUTIERREZ, J. Qualitative research methods and techniques in sciences

sociales. Madrid, Synthesis, 1994.

The research process in education

FRANCO, M.L.P.B. What is content analysis. Education Psychology Journals. São Paulo,
PUCSP (7): 1-31, Aug 1986.

GRAWITZ, M. Social Science Methods.

KRIPPENDORFF, K. Content Analysis Methodology: Theory and Practice. Barcelona, Editions

Paidós, 1990.

Naturalistic Inquiry
MINAYO, M. C. The challenge of knowledge: qualitative research in health. Rio de Janeiro, Hucitec.
Abrasco, 1992.

MORAES, R. Content Analysis: limits and possibilities. In: ENGERS, M.E.A. (Org).
Paradigms and research methodologies in education. Porto Alegre, EDIPUCRS, 1994.

MUCCHIELLI, R. Content analysis of documents and communications. Paris, The Bookstores

Techniques, 1977

OLABUENAGA, J.I. R.; ISPIZUA, M.A. The decoding of everyday life: methods of
qualitative research. Bilbao, University of Deusto, 1989.
PAGÉS, M.; BONETTI, M.; GAULEJAC, V.; DESCENDRE, D. The power of organizations. São
Paulo, Atlas, 1990.
RICHARDSON, R.J. Social research: method and techniques. São Paulo, Atlas, 1989.

ROSEMBERG, F. From intimacy to quiprocós: a discussion around content analysis.

Notebook CERU, São Paulo, (16):69-80, 1981.

SIMÕES, S. P. Meaning and possibilities of content analysis. Educational technology. Vol. 20

(102/103): 54-57, sep./dec., 1991.

Prof. Roque Moraes, PhD in Education. Faculty member at PUC/RS.

The provided text is a URL and does not contain translatable content. 12/13
31/01/2015 CONTENT ANALYSIS

The provided text is a URL and does not contain translatable content. 13/13