Anda di halaman 1dari 7

Content Analysis:

Used in qualitative research in Business as thematic – for the analysis of words and

Objective, systematic description of the content of text. Determines the presence of

certain textual items such as words, phrases etc that are repeated and thus can be
quantified or ‘themed’.

Stemler S., (2001) ‘An overview of Content Analysis’ Practical Assessment, Research
& Evaluation retrieved 15th Mar 2005

Content analysis is a reliable technique that enables researchers to sift and

categorise large volumes of data in a systematic fashion. It permits of inferences that
can be later triangulated with other methods of data collection. In particular it
examines patterns and trends in documents. It is considered a useful technique for
establishing themes and concepts that are repeated. The results are then an
objective assessment of inferences about the data messages and can identify
intentions, biases, prejudices and oversights amongst other themes (Berelson 1952).
Berelson identifies the following usages of content analysis:
• To reveal international differences in communication content;
• Detect propaganda;
• Identification of intentions or communication trends;
• Describing attitudinal and behavioural responses;
• Determining psychological or emotional responses / states (of mind).

The simplest form of content analysis consists of a word frequency count. This
assumes that the most frequently mentioned word(s) is the matter of greatest
concern. However, when considering a word frequency count, it is important to
consider the use of synonyms. Many documents will use synonyms for stylistic
reasons which may lead the frequency of a word to be underestimated. In addition, a
word may not represent a category adequately or a word may have multiple
meanings. Thus word frequency counts are often considered too simplistic.
However, the use of a word count may be used to formulate ideas or categories of
potential further interest / exploration.

It is the use of categorisation that gives the method of content analysis its strength.
Weber (1990 p37) says:
“A category is a group of words with similar meaning or connotations” but one must
also consider that “categories must be mutually exclusive and exhaustive” (GAO
1996 p37).

Categories can also be considered as concepts. Thus conceptual analysis is the most
common usage of content analysis. The initial research questions will generate the
samples which require analysis and categorisation. This can be generated from both
primary and/or secondary data sources. Concepts can be analysed both positively
and negatively - thus the researcher will consider whether the data describes the
concept in a positive manner or is opposed to the concept (or argument). The
outcome will be an analysis which identifies whether the data is more strongly in
favour or against a concept.
Alternatively, the content can be analysed through relational concepts. Thus the
researcher will be looking for linkages amongst the concepts. This will enable an
initial taxonomy to be coded into family trees and hierarchies of concepts can be

Two ways of coding are generally accepted. The first, emergent, requires the
researcher to read suitable extracts of the data and then to formulate initial an initial
coding frame or taxonomy (categorisation). Ideally this will be done with multiple
persons consulting and categorising the data, comparing results and consolidating
the categories. Categories are checked at intervals as the data is coded in order to
ensure completion (of categories) and validity of coding. The second is a priori
coding. In the latter way of coding, categories are established before the data is
analysed through agreement amongst the researchers and is revised as necessary
during the coding process. Categories must be shown, whichever method of coding
is undertaken, to be exhaustive and exclusive - that is that they do not overlap and
that they contain all the related data.

Coding units are required to establish the boundaries of each data set. In some
instances, if the data document is short enough e.g. a poem, the unit may be the
complete data item. A more common way might be to use the normal breaks in the
documents e.g. where the paragraphs are numbered, each section would be a unit,
or a sentence of paragraph or the actual paragraph itself, could also be a data unit.

The essential requirement of content analysis categorisation is that the classification

process can be shown to be reliable and consistent. This requires the rules of coding
to be made explicit and to be able to be shown to be stable and reproducible.

The advantages of content analysis are many and varied. The following are the main
arguments proposed (Busha and Harter 1980; Palmquist accessed 2005):
• Looks directly at data texts and hence surveys social interaction;
• Allows for both qualitative and quantitative analysis;
• Provides historical and cultural trends;
• Can statistically analyse text while at the same time uses the same data for
categorisation and concept analysis;
• Is an unobtrusive manner of analysing (social) interactions;
• Provides insight into human thought models and language use;
• Provides a (relatively) ‘exact’ analysis - because it is evidentially based.

Berelson B., (1952) Content analysis in Communication research New York: Free
Busha CH & Harter SP., (1980) Research Methods in Librarianship - Techniques and
Interpretation New York: Academic Press
GAO (US General Accounting Office) (1996) Content Analysis: a methodology for
structuring and analysing written material GAO/PEMD-10.3.1 Washington DC
Palmquist M., (2005)
Also at
Weber RP., (1990) Basic content analysis Newbury Park CA

1. Craig, R. T. (1981). "Generalization of Scott's Index of Intercoder Agreement." Public Opinion
Quarterly 45(2 (Summer)): 260-264.
Coherent analysis of open-ended survey questions or other verbal materials frequently makes
use of nominal
measurement, coding content units in terms of an unordered scheme of mutually exclusive and
exhaustive categories. Because the reliability of human coders and category schemes is variable,
coding reliability must usually be assessed in content analysis studies. In 1955, Scott proposed an
index of agreement between 2 coders which takes into consideration both the observed proportion of
agreement and the proportion that would be expected by chance. A content analysis study was done
which involved a total of 2,867 coding decisions using a 5-category coding scheme. Since the coding
task was difficult, all decisions were made by 3 independent coders, and 2/3 majority rule was used to
determine all final coding decisions. It appears that a generalization of Scott's proposal is applicable.

2. Hughes, M. A. and D. E. Garrett (1990). "Intercoder Reliability Estimation Approaches in

Marketing: A Generalizability Theory Framework for Quantitative Data." Journal of Marketing
Research 27(2 (May)): 185-195.
A review of intercoder reliability measurement in leading marketing journals reveals that most
marketing researchers are using percentage agreement, which is widely regarded as being
inadequate. For quantitative data, it is recommended that reliability estimation approaches
based on generalizability theory be used because of their superiority over approaches found in
classical test theory. The classical test theory framework does not explicitly take into
account the decision context and does not distinguish between absolute and relative error. Also,
generalizability theory estimators are chance corrected, reflect judges' systematic coding
errors, and are interpretable as interclass correlations. Hence, the generalizability theory approach
meets the criteria for appropriate intercoder reliability estimation that are widely accepted by
knowledgeable content analysis researchers. Winer's (1971) formulas and coefficient alpha are
evaluated within a generalizability theory framework.

3. Insch, G. S., J. E. Moore, et al. (1997). "Content Analysis in Leadership Research: Examples,
Procedures, and Suggestions for Future Use." Leadership Quarterly 8: 1-25.

4. Krippendorff, K. (1980). Content Analysis: An Introduction to Its Methodology. Beverly Hills, CA,
Sage Publications.

5. AU Tousignant, M Mishara, BL Caillaud, A Fortin, V St-Laurent, D The impact of media

coverage of the suicide of a well-known Quebec reporter: the case of Gaetan Girouard
Evidence of a media impact on suicide is mixed and needs further research. The main objective of
this article is to document the effects of the media coverage following the suicide of a well-known
and popular television reporter in Quebec, Canada. A content analysis of the printed media and an
analysis of suicide rates during the following year, of coroners' records and of calls to Suicide
Prevention Centres during the following 3 months was conducted. Most guidelines for responsible
reporting of a suicide were not applied. The results showed a rise in the suicides rates immediately
after the reporter's suicide, especially by hanging as in the original case. A cluster of six suicides by
hanging also took place in the small municipality where the reporter's suicide occurred. There was
also an indication of direct influence in the coroners' records and a rise in calls to Suicide Prevention
Centres. This research indicates that the reporting of the suicide of a popular figure preceded an
important rise in the number of suicides. A possible theoretical explanation is that a positive role
model appeared to suddenly fail to cope with
life, thus creating high distress and cognitive dissonance in the audience. The news media should
apply more caution and follow recommended guidelines in reporting this type of news 1919- 1926
JI Soc. Sci. Med. PY 2005 PD MAY VL 60 IS 9

1. Groot, MM Vernooij-Dassen, MJFJ Crul, BJP Grol, RPTM TI General practitioners (GPs) and
palliative care: perceived tasks and barriers in daily practice
Background: General practitioners (GPs) play a crucial part in palliative care. The quality of care can
be improved by investigating and addressing barriers perceived by GPs in daily practice. The aim
of this study was to investigate GPs' task perception and barriers involved in palliative care. Methods:
Qualitative focus group study. We gathered together a group of GPs representing a broad range of
experience in palliative care. Content analysis was performed to derive a comprehensive view of
tasks and barriers in daily palliative care. Results: GPs described their palliative care tasks as
satisfactory and varied, but burdensome. Palliative care tasks included somatic and psychosocial
care. Opinions differed with respect to whether the coordination of care belonged to the primary GP
tasks. Barriers were classified according to three levels: (1) personal: barriers related to knowledge,
skills, emotions; (2) relational: barriers concerning communication and collaboration; (3)
organizational: barriers related to the organization of care and compartmentalization in healthcare.
Conclusions: This study revealed a complex web of tasks and barriers. It may be possible to trace
back a problem (lack of knowledge, for example) on the personal level to an isolated knowledge
gap, but the problem may well have originated from communication or compartmentalization
problems. To maintain GPs' feeling of being at ease with palliative care requires helping them
acquire the appropriate balance between technical and organizational interventions and a
compassionate orientation to their terminally ill patients.
111 - 118 Palliat. Med. 2005 19 2

2. Halek, K Murdoch, M Fortier, L Spontaneous reports of emotional upset and health care
utilization among veterans with posttraumatic stress disorder after receiving a potentially
upsetting survey
This study assessed the effects of a potentially distressing mailed survey on the emotional well-being
and health care utilization (HCU) of 4,918 male and female veterans who applied for posttraumatic
stress disorder disability benefits. Content analysis of spontaneous comments, in combination with
of subjects' HCU before and after receipt of the survey, suggested that spontaneously disclosed
episodes of emotional upset were rare. In general, surveyed veterans' HCU decreased after receipt
of the survey.
142 - 151 JI Am. J. Orthopsychiatr. 2005 JAN 75 1

3. Maguire, B American professional wrestling: Evolution, content, and popular appeal

Over the past several years, professional wrestling, now referred to as sports entertainment, has
become a hugely popular cultural phenomenon. There are several reasons to account for why tens
of millions of Americans are attracted to this form of entertainment, but this article centers attention
on three allures that stand out: excitement, intrigue, and political incorrectness. Content analysis
suggests that these three foci form the core of pro wrestling program content. The present paper
identifies the macrosocial forces that explain sociologically why these themes are especially
today: community breakdown, social disenchantment, and political correctness.
155 - 176 JI Sociol. Spectr. 2005 MAR-APR 25 2

4. Monnickendam, M Savaya, R Waysman, M Thinking processes in social workers' use of a

clinical decision support system: A qualitative study
The authors examined the thinking processes in the use of a decision support system (DSS) by
social workers in a human services agency to determine whether they used the system to improve
their case reasoning. Information was obtained from in- depth interviews with eight social workers
who used a DSS in their work and from content analysis of "narrative justifications" appended to
1,074 decisions that differed from those recommended by the DSS. Findings show that the social
workers used the DSS in a perfunctory manner in typical cases, but as an aid to thinking and
reflection in atypical cases, in which they were uncertain of how to decide. On the whole, the social
workers showed little interest in the DSSs recommendations, but reported that the process of
entering data and answering the model's questions about the case were useful when the case was
atypical. The authors suggest that computer support systems should be designed to help workers
think through atypical cases.
21 - 30 JI Soc. Work Res. 2005 MAR 29 1

5. Wansink, B Westgren, RE Cheney, MM Hierarchy of nutritional knowledge that relates to the

consumption of a functional food
Objective: We assessed how consumption of a functional food relates to different combinations of
nutritional knowledge. Methods: American and Canadian subjects were asked by mail survey about
their level of knowledge about soy and were assigned to one of four groups based on whether they
had 1) no
knowledge of soy, 2) attribute-related knowledge of soy, 3) consequence-related knowledge of
consuming soy, or 4) both types of knowledge. Content analysis and analysis of variance were
performed. Results: The level of nutritional knowledge about soy did not necessarily influence how
much people liked soy but was related to how much people consumed soy. In particular,
consumers who were able to link attribute-related knowledge about soy to consequence-related
knowledge about consuming soy were much more likely to consume soy than were those who only
had one type of knowledge (average P < 0.007). Conclusions: Nutritional knowledge most likely
correlates with consumption when people have attribute-related knowledge of the food and
consequence-related knowledge of how it will benefit them. It is not the amount but the type of
knowledge that matters. Educational strategies based only on attribute-related knowledge of
functional foods and healthy products ("passing the nutrition quiz") may not effectively encourage
the actual consumption of the food. Health care professionals and
dietitians must link food attributes with personal health consequences when communicating to
their patients. (c) 2005
264 - 268 JI Nutrition 2005 FEB 21 2

6. Foote, SB Halpern, R Wholey, DR Variation in Medicare's local coverage policies: Content

analysis of local medical review policies
Objective: To assess variation in the content of Medicare's local medical review policies. Study
Design: Six case studies to compare differences in coverage policies by diagnosis codes, procedure
codes, and indications for use. Methods: All carrier policies from 48 carrier contracts (n = 52 13)
posted to the
Centers for Medicare & Medicaid Services Web site were downloaded on May 31, 2001. All policies
in the data set were coded based on a typology: new technology (NT), extensions of new
technology (TE), and utilization management (UM) of widely used procedures. We identified policies
addressing the same procedure or technology. We required at least 20 separate policies in each
case study to allow meaningful comparisons. We randomly selected I case study of a diagnostic and I
for a treatment modality from each policy type (NT, TE, and UM). Results: Given previous research
on local carriers, we expected to find variations among policies in each case study. We found
substantial similarity, however, among policies covering the NT and TE types. We found significantly
more variation among our UM-type case studies. Conclusions: Medicare legislation has called for
greater coverage policy consistency in Medicare. This analysis on variation in policy content, part of
a larger study on variation in Medicare's local coverage process, provides data on policy content
differences. Policy reform should reflect the nature of and reasons for policy variation as suggested
by the findings of this research.
181 - 187 JI Am. J. Manag. Care 2005 MAR 11 3

7. Luthy, C Cedraschi, C Perrin, E Allaz, AF How do patients define "good" and "bad" doctors?
- Qualitative approach to the representations of hospital patients
Questions under study: Knowledge of hospital patients' perceptions of doctors' qualities is limited.
The purpose of this study was to explore hospital patients' definitions of "good" and "bad" doctors.
Methods: Semi-structured interviews conducted with 68 consecutive hospital patients. The questions
explored the characteristics of good/bad doctors. Responses were subjected to content analysis.
Results: The patients' mean age was 72.7 (+/-15) years; 61% were female. Content analysis
produced 9 categories connoted positively/negatively; the mean number of categories/patient
response was 2.4 (+/-13), ranging from 1-6. Sensitivity/ insensitivity to feelings were in the
forefront, together with the importance of the relational dimension and the need to provide
treatment tailored to the patient's needs. Patients' responses emphasised "bad" doctors' use of
medicine as selfserving and not serving the patient. Conclusion: This qualitative enquiry made it
possible to gather information on the patients' expectations or beliefs outside physicians' or health
researchers' pre-established categories. It emphasised that acknowledging possible areas of
uncertainty may be less threatening for the doctor's image than exhibiting scientific proficiency
unadapted to the patient's expectations and needs.
82 - 86 JI Swiss Med. Wkly. 2005 FEB 5 135 5-6

8. Hudson, S Miller, GA The responsible marketing of tourism: the case of Canadian Mountain
Heli-tourism represents one of the great dilemmas and conflicts between recreational enjoyment of
the wilderness and the conservation of the fragile alpine and mountain areas where the activity
takes place. The question of responsibility towards the environment is one, which tourism operators
generally seem reluctant to accept but one operator that appears to have taken a proactive
approach to environmental issues in mountain regions is heli-operator Canadian Mountain Holidays
(CMH). This paper expands on the limited amount of research that exists on the complex
relationship between tourism and the environment by applying a responsible marketing model to
CMH. This model is grounded on previous literature in marketing, and strategic and environmental
management. Interviews with key stakeholders, observational research, and content analysis of
communication materials, were used to identify how near CMH is to finding a balance between
responsible action and the communication of these activities. 133 - 142 JI Tourism Manage. 2005
APR 26 2

9. Wasko, MM Faraj, S Why should I share? Examining social capital and knowledge
contribution in electronic networks of practice
Electronic networks of practice are computer-mediated discussion forums focused on problems of
practice that enable individuals to exchange advice and ideas with others based on common
interests. However, why individuals help strangers in these electronic networks is not well
understood: there is no immediate benefit to the contributor, and free-riders are able to acquire
the same knowledge as everyone else. To understand this paradox, we apply theories of collective
action to examine how individual motivations and social capital influence knowledge contribution in
electronic networks. This study reports on the activities of one electronic network supporting a
professional legal association. Using archival, network, survey, and content analysis data, we
empirically test a model of knowledge contribution. We find that people contribute their knowledge
when they perceive that it enhances their professional reputations, when they have the experience
to share, and when they are structurally embedded in the network. Surprisingly, contributions
occur without regard to expectations of reciprocity from others or high levels of commitment to the
35 - 57 JI MIS Q. 2005 MAR 29 1

10.Beck, CT Benefits of participating in Internet interviews: Women helping women

Advantages of face-to-face qualitative interviews for participants have been addressed in the
literature. The benefits of participating in qualitative interviews over the Internet, however, have
yet to be discussed. Based on the experiences of 40 women who made up the sample for an Internet
study on birth trauma, the author describes in this article the benefits reaped by these mothers
through their participation in qualitative e-mail interviews. She used content analysis to identify
these benefits. Seven themes emerged: experiencing caring by being listened to and acknowledged,
sense of belonging, making sense of it all, letting go, being empowered, women helping women,
and providing a voice. 411 - 422 JI Qual. Health Res. 2005 MAR 15 3

11.Lee, TT Nurses' concerns about using information systems: analysis of comments on a

computerized nursing care plan system in Taiwan
Aims and objectives. The purpose of this study was to pursue a deeper understanding of nurses'
perceptions of a computerized nursing care plan as written in comments on a survey. Background.
Studies have examined nurses' perceptions of computer use, but comments written on surveys of
computerized nursing care planning have never been reported and the pattern and meaning of
such feedback have not been systematically analysed. Design. Of 738 nurses in Taiwan who
completed a two- part survey to evaluate a computerized nursing care plan, 202 responded on part
II to an open-ended question about overall perceptions of a nursing information system. Methods. A
sample of 10 questionnaires with comments was subjected to qualitative content analysis. Key
words and phrases were identified, then grouped into categories, which were used to code the
remaining respondents' comments. Another 20 were used to pilot test these coding categories. The
final coding categories were applied to all 202 questionnaires with comments for data analysis.
Results. Nurses were concerned about inconvenient access to computers; reduced work efficiency;
inability to individualize patient care and nursing specialty deficiencies; poor content design,
system function, and system integration; using the system as a policy requirement; and privacy and
legal issues. Conclusion. Hardware availability, content design and user training/education
programmes are critical issues that affect nurses' use of computers in their daily practice. Relevance
to clinical practice. These comments can be applied in strategic planning and staff development
programmes to further use of information systems in patient care.
344- 353 JI J. Clin. Nurs. 2005 MAR 14 3

12.Mulvey, MS Stern, BB Content analysis research themes 1977-2000: Evolution and

More than two decades ago, Hal Kassarjian introduced the content analysis methodology to
consumer researchers in his 1977 Journal of Consumer Research article "Content Analysis in
Consumer Research." Our research has two goals. First, from a historical perspective, we trace the
evolution in the use and application of content analysis in-field from 1977-2000. Second, we
analyze the substantive issues and thematic domains that dominate this body of research. By
integrating the set of studies in terms of themes, we are in a better position to describe current
knowledge and practice, evaluate theoretical progress, identify gaps and weak points that remain,
and plot a course for future research.