INDEPENDENT MEDIA

Media for Transparent Governance

Content analysis

What is content analysis
ANNEX 1
ANNEX 2
ANNEX 3
Content Analysis Principles and Methodologies

WHAT IS CONTENT ANALYSIS?

Content analysis is a research method that uses a set of procedures to make valid inferences from the text (Weber, 1990: 9)

Content analysis is a research technique for making replicable and valid inferences from the data to their context (Krippendorf, 1980: 21)

Content analysis is any technique for making inferences by objectively and systematically identifying characteristics of messages (Holsti, 1969: 14)

These definitions point to the following features of content analysis:

1) it is a research method/research technique – this means that content analysis is systematic, objective, and replicable

2) the analysis focuses on the message (text) – the message/text is anything written, visual, or spoken that serves as a medium for communication such as books, newspaper or magazine articles, speeches, official documents, pictures, paintings, TV programs, radio scripts, films

3) one of its goals is to describe the message/text – the description can be done in three ways:

3.1. identifying the content (e.g., topics/issues raised, key actors/personalities involved, characteristics of the key actors/personalities, treatment of the issues/personalities, etc.) and structure (format of presentation, length of the material, size of the materials, fonts used, visuals used, sound effects used, etc.) of the message
3.2. evaluating the quality of the message in terms of accuracy, completeness, understandability, appropriateness to the target audience
3.3. making comparisons among various message sources and/or across time periods

4) its other goal is making inferences about the:

4.1. probable character of the message sources
4.2. probable character of the audiences
4.3. possible effects of the messages on the audiences
4.4. the cultural norms and social behaviors that a message appears to reflect

To do content analysis study on media messages on corruption means:
1. Basically, to find out how the media has covered particular cases of corruption;
2. Specifically:
2.1. to describe the content and structure of the reports/features on corruption – what specific aspects of the corruption case were covered; which institutions, groups or persons were mentioned in relation to the case; whether these institutions, groups or persons were treated positively or negatively in relation to the corruption case; how much space or airtime was devoted to the case, etc.
2.2. to evaluate the coverage of the corruption case – whether it is accurate, complete, balanced, understandable, etc.
2.3. if applicable, to determine the similarities and differences in the way different media sources (e.g., one newspaper vs. another newspaper; newspapers vs. radio) covered the corruption issue; or in the way the corruption issue was covered across a specific time period (on weekly, monthly, or yearly basis)
2.4. to present a sound, data-based assessment of the performance of the media in its coverage of the corruption issue (its good and bad points, strengths and weaknesses in relation to the coverage of the issue) and of the possible reasons behind this performance (for example, was the poor performance due to lack of journalistic skills, fear of government sanctions, the commercial orientation of the owners of the media business, etc.)
2.5. to discuss the possible impact of the coverage on the audiences (did the coverage help them to understand the issue better, to make an intelligent stand on the issue, to appreciate the relevance of the issue on their own personal lives, etc.)
2.6. to offer some recommendations on how the media’s coverage of corruption issues may be improved.

WHAT ARE THE TYPES OF CONTENT ANALYSIS STUDIES?

Content analysis studies make be classified in several ways:

1. According to the medium analyzed – whether:
- newspapers
- radio
- TV
- film, etc.
- or a combination of media

2. According to the purpose of the analysis – whether:
- content classification – determining themes, topics, treatment, etc.
- content interpretation – making inferences about the message source, recipients, impacts
- or a combination of these two

3. According to the type of data involved
- quantitative or numerical – uses statistical data/tables
- qualitative or textual – cites sentences/paragraphs from the documents analyzed
- or both

These classifications are not mutually exclusive. Content analysis studies, in fact, combine these three forms of classification. Thus, for example:
When a researcher conducts a content analysis of media’s coverage of corruption issues, this study can be:

- a content analysis of one newspaper only, undertake content classification only, and use numerical data only to analyze and present the findings; or
- a content analysis of several newspapers that looks at content classification only and uses numerical data only; the study will describe each paper’s coverage of the corruption issue, as well as compare the papers (their similarities and differences) in terms of how they covered the issue; or
- a content analysis of two or more media types (e.g., newspaper, radio, TV) that will undertake both content classification and content interpretation, but presenting numerical data only; or
- a content analysis of two or more media types that will have both content classification and content interpretation, and will use both quantitative and qualitative data; or any other combinations on items 1 – 3 above.

The researcher decides on the approach most suitable for his/her research depending on the objectives of the study and the resources available. Obviously, however, studies that focus on more media, combine content classification and interpretation, and use both numerical and textual data yield more information and insights than those that focus on fewer media, undertake content classification only, and use only one type of data.

WHAT ARE THE PURPOSES AND APPLICATIONS OF CONTENT ANALYSIS?

Content analysis can help determine:

1. the image of a person, group or institution as portrayed by the various media
Examples:
The newspapers’ portrayal of the president during his first month in office Television’s portrayal of the senatorial candidates in the last national elections The image of women in television advertisements The image of men in romance films Etc.

2. the image that a person or an institution projects through the messages disseminated
Examples:
The self-presentation of the senatorial candidates in their campaign speeches The image of Company XYZ as seen through its public relations campaign materials Drawing up a profile of the youth based on their autobiographies Etc.

3. how an issue is covered/treated by the media
Examples:
A content analysis of the print and broadcast media’s coverage of the national elections Time magazine’s coverage of the US-Iraq conflict The coverage of the death penalty issue in television talk shows Etc.

4. what specific issues/concerns are important for the media/public
Examples:
Setting the agenda: front page news in Newspaper ABC in the last five years
A content analysis of the research priorities of academic journals on communication during period 2000 – 2003 The youth’s expectations from the government: a content analysis of school children’s letters to the president Etc.

5. if there are differences in the way different media portray a person, group or institution; or in the way the different media covered an issue
Examples:
A comparative analysis of the image of women in Japanese and Korean television soap operas A comparative content analysis of the coverage of the national elections in the three leading national newspapers Etc.

6. if media’s portrayal of a person, group or institution has changed over time
Examples:
Trends in the print media’s coverage of the University of the Philippines from 1908 to 2000 Media’s coverage of the presidential candidates before and after the elections A comparative analysis of the coverage of women’s issues in community newspapers in the last three decades

WHAT ARE THE STEPS INVOLVED IN CONDUCTING CONTENT ANALYSIS?

A content analysis of media’s coverage of corruption issues can, therefore, yield important information about several things:

1. how the different persons and institutions involved in the issue are covered by the media – who are portrayed favorably and who are portrayed negatively

2. what aspects of the corruption issue are given more importance by the media – the facts of the case, the personalities involved, the public’s sentiments, the larger implications of the case, etc.

3. if there are differences in the way different media cover the corruption issues

4. if the coverage of the corruption issues has changed over time, and what these changes are

Content analysis involves the following steps:
1. Identifying the content analysis topic, problem and objectives
2. Identifying the content analysis variables and measures
3. Formulating the content analysis instrument
4. Pretesting the content analysis instrument and determining inter-coder reliability
5. Selecting the sample for content analysis
6. Data gathering in content analysis
7. Procedures for processing and analyzing content analysis data
8. Interpreting content analysis findings
9. Writing the content analysis report

IDENTIFYING THE TOPIC, PROBLEM, AND OBJECTIVES
The researcher should define the scope and significance of the study. Ask yourself the following questions:

1. What issue or issues will I analyze?
2. What media sources will I look at? (i.e., consider if you want one or several media types)
3. What are my specific research goals?
- Will I only describe the content of the messages, or will I make inferences?
- If inferences are to be made, will these be about the source, audience or impact of the messages?
- Will I look at the similarities and differences in the coverage of the issue a) across media types and/or b) over time?
- Are there perspectives or hypotheses that I want to test? (for example, a researcher can assess media practitioners’ compliance to ethical and legal standards of news coverage in the way they handle a particular issue)
- Do I want to make recommendations based on the results? For whom are these recommendations?
4. Why is it important to do this study? What practical and theoretical benefits will it bring? For whom will the study be useful?
5. Do I have the resources (time, money, manpower, etc.) to do this study?

EXAMPLE
Issue: Impeachment trial of President Joseph Estrada
Research problem: How did the three leading national broadsheets cover the impeachment trial of President Estrada?

Objectives:
1. To describe how each broadsheet covered the impeachment trial in terms of the following:
1.1. prominence (importance) given to the case
1.2. specific aspects of the impeachment case that were discussed
1.3. specific aspects of the impeachment trial that were discussed
1.4. treatment of the personalities and institutions involved in the impeachment trial
2. To identify similarities and differences in the broadsheets’ coverage of the impeachment trial
3. To present recommendations as to how the broadsheets’ coverage of similar issues may be improved.

IDENTIFYING THE CONTENT ANALYSIS VARIABLES AND THEIR MEASURES

The choice of the variables to include in the content analysis study is guided by the problem and objectives of the study. It is also helpful to consult other studies similar to your own research to get ideas about what variables should be included in the analysis. Basic information about the materials to be analyzed should, however, always be included in the list of variables. These are:

1. Title of the material (e.g., title of the newspaper article, title of radio program, etc.
2. When the material was published/produced (e.g., date of newspaper issue, date of radio broadcast, date of release of the film, date of publication)
3. For newspaper articles, radio programs, TV programs, online materials: Where the material was published/broadcast (i.e., name of newspaper, radio station, TV station, URL)
4. For other print materials (magazines, brochures, journals, etc.), films, works of art: Who produced the material (name of publisher, producer, artist)
5. Length of the material (in column inches, number of seconds or minutes)
6. (If the materials are of different formats) Format of the material (e.g., news article, editorial, column article, jingle, radio/TV spot) or type of material/program where the message came from (e.g., brochure, poster, talk show, soap opera, news program); for materials like films, the type of film is identified (e.g., romance, comedy, action, horror, etc.)

Basic information about the coding process (coding is the task of filling up the content analysis instrument) should also be included, namely:

1. Name of coder
2. Date of coding
3. Time started
4. Time ended

As mentioned, the main content analysis variables should be chosen guided by the objectives of the research, and measures of these variables should be formulated. This means going the following steps:

1. Listing the variables, as determined by the problem statement and research objectives
2. Developing operational definitions of the variables – operational definitions yield the categories of the variables and how to determine into which category of a particular variable a particular content belongs.

Example:
Variable: Treatment of President Estrada in the newspaper article

Operational definition: This refers to whether the portrayal of President Estrada is positive, neutral or negative.
- Treatment is positive if there are more statements favorable to the president than negative ones
- Treatment is negative if there are more statements unfavorable to the president than positive ones
- Treatment is neutral if there is an equal number of statements favorable and unfavorable to the president

EXAMPLE
Research problem: How did the three leading national broadsheets cover the impeachment trial of President Estrada?

Variables and measures

1. Prominence given to the impeachment case – this refers to the importance the broadsheets gave to the case. Importance is to be gauged from the following:
1.1. Page location of the article – whether front, inside or back
1.2. If frontpage, whether the article is headline news or not
1.3. If frontpage article is not headline news, the location of the article on the front page – whether above the fold or below the fold
1.4. Amount of space given to the article – measure in terms of column inches
Importance will also also be determined by looking at:
1.5. The total number of articles on impeachment vs. the total number of articles in the newspaper issue analyzed

2. Speficic aspects of the impeachement case that were discussed – this refers to the impeachment-related topics that were discussed in the articles. These topics include:
2.1. the accusations against the president (the basis for filing the impeachment case)
2.2. the president’s response to the accusations
2.3. the legal procedures for filing the case
2.4. the public’s opinion about the accusations
2.5. others

3. Specific aspects of the impeachment trial that were discussed – this refers to the specific topics about the impeachment trial, other than the case itself, that were discussed in the articles. These topics include:
3.1. the trial proceedings – report of activities about the trial itself
3.2. the role of the Senate in the impeachment trial
3.3. the role of the judiciary in the impeachment trial
3.4. the role of the media in the impeachment trial
3.5. the people’s sentiments about the impeachment trial
3.6. others

4. Personalities and institutions involved in the impeachment trial – these include the president, his lawyers, the senators, the media, the socio-civic groups, the president’s friends, etc.

5. Treatment of the personalities involved in the impeachment case and/or impeachment trial – this refers to whether the portrayal of a person or institution is favorable, neutral, or unfavorable.
5.1. favorable – if there are more positive than negative statements about the person/institution
5.2. neutral – if there is an equal number of positive and negative statements
5.3. unfavorable – if there more negative than positive statements

FORMULATING THE CONTENT ANALYSIS INSTRUMENT

The content analysis instrument consists of two parts:

1. the content analysis form – the main research instrument where the coders will enter the information about the materials and messages analyzed
2. the content analysis codebook – contains the definitions of the variables and of the categories under each variable; the codebook serves as the coder’s guide when filling up the content analysis form.

Annexes 1 and 2 contain samples of the content analysis form and the content analysis codebook, respectively.

PRETESTING THE CONTENT ANALYSIS INSTRUMENT AND
DETERMINING INTER-CODER AGREEMENT

It is important to pretest the content analysis instrument for several reasons:

1) to determine if your variables are appropriate and sufficient for the materials and messages to be analyzed
2) to determine if your variable definitions and coding instructions are clear, i.e., easy for your coders to understand and follow
3) to identify possible problems that will be encountered in the actual coding of the messages – for example, is it possible to complete the data gathering within the schedule you are targeting? Is the amount you are going to pay the coders reasonable and commensurate to the amount of work that they will do?

The most important goal of pretesting is to assess the quality of your content analysis instrument. The main test of quality is the extent to which several coders code the same message in the same way – this is called inter-coder agreement, and may be measured informally or statistically through the computation of the inter-coder reliability coefficient.

The following are the steps for pretesting the instrument and assessing inter-coder agreement:
1. Select one article to be content analyzed
2. Ask several coders to code this article using the content analysis form and guided by the content analysis coding instructions. The coders should conduct their coding independently; they should not consult with each other or the principal researcher. No clarifications should be given; the coders should only rely on how they understand the instructions as written on the codebook.
3. After all coders have finished coding, go through each of the items in the content analysis form and compare the coders’ entries for each item. Pay particular attention to differences in the coding – find out and resolve the differences in the coders’ interpretation of the coding instructions and/or the difficulties they encountered in analyzing the messages.
4. The “sharing of experiences” in the coding of the pretest material may be used as a gauge of inter-coder agreement. If there are few differences in the way the coders coded the message, and if there are few items where they differ in the interpretation of the instructions, then one can say that there is high inter-coder agreement on the application of the content analysis instrument.
5. A more formal way of assessing inter-coder agreement is to compute the composite reliability coefficient using the formula (Holsti, 1969):

Composite reliability =
N (average inter-judge agreement)
1 + [(N – 1) (average inter-judge agreement)]

The value of the composite reliability coefficient ranges from 0 to 1. A coefficient of 0.80 or higher indicates high inter-coder agreement. If the computed coefficient is lower than .80, another pretest should be done – after clarifying the items where the coders differed in their coding – using a different message/material (not the same one that was used in the first pretest).

SELECTING THE SAMPLE FOR THE CONTENT ANALYSIS

After the content analysis instrument is finalized, the researcher must select the materials that will constitute his/her sample. In line with this, the researcher can be guided by these questions:

1. What types of media materials should I include? – This is defined by the research problem and objectives. In our example, it is clear that only newspapers will be content analyzed; further, only the three leading national broadsheets will be covered by the study.

2. What time period should I cover? – Again, depending on the research problem and objectives, the time period may be long (decades, years) or short (days, weeks, months). In the example given, the impeachment trial lasted only two months; thus, only the newspaper issues in those two months will be considered for the analysis.

3. Will I get all materials within the time period specified, or will I select a sample of materials only? – If the time period is long and/or there are too many types of materials to be analyzed, sampling is recommended; if the time period is short, sampling may not be necessary. Availability of resources for content analysis and access to the materials are additional considerations – i.e., if resources are limited and/or it is difficult to access the materials, it is advisable to do sampling). In our example, because only two months and three broadsheets are involved, and because there are resources for the study and the newspaper issues are easy to find, we can opt to include all the issues (62 issues/newspaper x 3 newspapers = 186 issues) in the content analysis.

4. If only a sample of materials is to be used, how will I select the sample? – A representative sample is recommended, because it allows making generalizations from the sample to the population.

DATA GATHERING
This is the stage where the coders code all the materials/messages using the final content analysis instrument. Prior to actual data gathering, the lead researcher must conduct a training on the coding procedures.

PROCESSING AND ANALYZING THE DATA
The data may be processed and analyzed manually or using a computer software. Obviously, the latter is the better option. Data processing entails the following:

- Formulating the coding guide – this somewhat similar to the content analysis coding instructions, but the focus is on how to encode the data found in the content analysis forms into the computer data file. (See Annex 3 for a sample coding guide).
- Constructing the data file
- Encoding the data into the data file

After the data have been encoded, the researcher can proceed with the data analysis. This entails:

- generating the frequency tables
- generating cross-tabulations showing the desired comparisons
- if there are hypotheses to be tested, running the appropriate statistical tests

INTERPRETING THE FINDINGS

Interpretation means “making sense” of the results of the data analysis. This is the stage where the researcher examines the data to find the answers to the research problem and objectives. Depending on the objectives of the study, the interpretation will include one or all of the following:

- describing the characteristics of the materials/messages analyzed
- discussing the similarities and differences among the materials analyzed in terms of the messages they disseminated
- discussing similarities and differences in the messages over time
- evaluating the messages vis-а-vis certain standards, e.g., code of ethics, legal standards, editorial policies
- making inferences about the characteristics of the source of the messages
- making inferences about the characteristics of the audiences of the messages
- making inferences about the probable impacts of the messages on the audience
- making inferences about the social context (history, culture, norms, politics, government, etc.) within which the messages are produced and disseminated
- formulating recommendations

In broader terms, the tasks of the researcher during the data interpretation stage are 1) to make the numbers understandable to the readers, 2) to provide answers to the research problem and objectives, 3) to offer possible explanations for the findings and their implications for the different sectors of society.

WRITING THE RESEARCH REPORT

Like any other research report, a content analysis report follows the rules and conventions of good writing. Just like any research report, too, a content analysis report must have a clear, comprehensive, and concise discussion of the research problem and its significance, the research objectives, the theoretical perspectives guiding the study, the methodology, the findings of the study and their implications, and the study’s conclusions and recommendations.

CONTENT ANALYSIS: SOME GENERAL POINTS TO REMEMBER

Content analysis is one of the most popular methods employed in communication/ media research. Among other things, it is one of the most useful ways of monitoring message sources; thus, it is one of the best ways of monitoring media performance, in such matters as:

- compliance with standards for crime reporting – e.g., the extent to which newspapers publish prejudicial pretrial information (Imrich, et. al., 1995; Umali, 1999)
- compliance with standards for reporting results of opinion polls, e.g., the extent to which newspapers include important information about the polling methodology (Hardmeier, 1999)
- compliance with editorial policies
- adequacy of coverage of important issues and sectors, e.g., the media monitoring project on the coverage of various corruption issues in Bulgaria (http://www.anticorruption.bg/eng/mmonitoring/case); studies on the Philippine media’s coverage of elections (CMFR, 2004; Umali et. al., 2001) and of children’s issues (Sevilla et. al., 2001), and their portrayal of women (Azarcon-dela Cruz, 1988).

Information gathered through content analysis is, therefore, one of the most useful ways by which media practitioners can arrive at a systematic and objective assessment of their strengths and weaknesses as far as the practice of their profession is concerned. Consequently, content analysis can help them identify the specific actions to take so that they can be more effective in their commitment to being the watchdogs of society.

As earlier mentioned, content analysis is also helpful in making inferences about the probable characteristics of the message sources and/or audiences, and the probable impact of the messages on their audience. Examples of such inferences are 1) the link between sensationalism in crime reporting and the desire for profit, 2) the inadequate coverage of an issue as an indicator of the journalists’ insufficient grounding in the issue, lack of basic reporting skills, laziness, etc, and 3) the media practitioners’ failure to observe ethical and legal standards as a consequence of their lack of knowledge of and/or respect for these standards.

Inferences about the probable characteristics of the audience take off mainly from the argument that the media are “only giving the people what they want”. However, such inference is often presented precisely to question the validity of the argument, or to pose it as a food for thought. Is it correct to say that news programs are sensational because people want sensationalism, not serious information? Is it reasonable to argue women’s magazines are trivial and escapist because that is how women are – trivial and escapist? Such inferences, then, are helpful hints for the media to rethink the way they regard their respective audiences.

Inferences about the characteristics of the audience can also be situated in the larger social context. Studies on the portrayal of women in the media are one of the best examples of this kind of inference-making. Thus, by presenting those inferences, content analysis studies reveal the subtle manifestations of power relations, class differences, gender biases, racist ideologies, etc. that have found their way in the media messages. Through such inferences, content analysis can thus help the media better understand how they contribute to the perpetuation of the status quo and how they may help institute changes in society.

Analyzing the probable impacts of the messages on the audience, on the other hand, are useful reminders to the media practitioners about their responsibilities to the people. It cannot be overemphasized that the media need to be circumspect about the kind of messages they disseminate, to be conscious of the power of words.

For their part, content analysis researchers should interpret their findings and draw inferences from these findings with restraint, objectivity, and sound reasoning. Their interpretations and inferences cannot be colored by personal biases and preferences. Doing so is a sure-fire way of ruining a good research, and of subverting the potential contributions of content analysis to the improvement of media and society.

References:
Babbie, E. (1998). The practice of social research. California: Wadsworth Publishing.
Holsti, O. (1969). Content analysis for the social sciences and humanities. Massachusetts: Addison-Wesley Publishing Co.
Krippendorf, K. (1980). Content analysis: An introduction to its methodology. California: Sage Publications.
Lombard, M. et. al. (2004). Practical resources for assessing and reporting intercoder reliability in content analysis research projects. http://www.temple.edu/mmc/reliability
Weber, R. (1990). Basic content analysis, 2nd. ed. California: Sage Publications.

Works cited:
Azarcon-dela Cruz, P. (1988). From virgin to vamp: Images of women in Philippine media. Manila: Asian Social Institute.
Center for Media Freedom and Responsibility (CMFR). (2004). Citizens’ Media Monitor: A report on the campaign and elections coverage in the Philippines 2004. Quezon City: CMFR.
Hardmeier, S. (1999). Political poll reporting in Swiss print media: Analysis and suggestions for quality improvement. International Journal of Public Opinion Research 11 (3): 257-274.
Imrich, D. et. al. (1995). Measuring the extent of prejudicial pretrial publicity in major American newspapers: A content analysis. Journal of Communication 45 (3): 94-117.
Sevilla, A. et. al. (2001). Reporting about children in the Philippines. In Goonasekara, A. (Ed.), Children in the news: Reporting of children’s issues in television and the press in Asia. Singapore: AMIC.
Umali, V. et al. (2001). Towards a discourse on political communication: The case of the 2001 Philippine national elections. Paper presented at the Annual Conference of the International Association for Media and Communication Research held in Barcelona, Spain on July 2001.
Umali, V. (1999). A comparative study of sex crime reporting in selected ASEAN newspapers. Unpublished paper for a masteral course.
Case studies from http://www.anticorruption.bg/eng/mmonitoring/case

ANNEX 1
Sample Content Analysis Form

Coverage of the Impeachment Trial of President Joseph Estrada
by the Three Leading National Broadsheets

Coder’s name: ____________________________________________________
Newspaper: ______________________________________________________
Date of issue: _____________________ Day (encircle): M/T/W/TH/FR/SA/SU
Article ____ of _____ articles Type of article: __________________
Time started:_____________________ Time ended: ___________________

1. Page location of the article (check): ___ Front ___ Inside ___ Back
2. Type of front page article: ___ Headline ___ Not headline ___ NAP
3. Location of front page article: ___ Above fold ___ Below fold ___ NAP
4. Length of article in column inches: ________________________________
5. Impeachment case topic/s discussed (check all that apply):
___ Accusations against Pres. Estrada (specify)
___ Pres. Estrada’s response to the accusations
___ Legal procedures for filing the case
___ Public’s opinion about the case
___ Others (specify): __________________________________________

5.1. Please give a brief summary of the content for each topic:

6. Impeachment trial topic/s discussed (check all that apply):
___ Trial proceedings
___ Role of the senate
___ Role of the judiciary
___ Role of the media
___ People’s sentiments about the trial
___ Others (specify): __________________________________________

7. Personalities/institutions mentioned in the article (list all major personalities/ institutions):
................................................................................................................
................................................................................................................
................................................................................................................
................................................................................................................

8. Treatment of the personalities/institutions:

9. Please write down below any other observations you might have that are relevant to the objectives of the study:
................................................................................................................
................................................................................................................
................................................................................................................
................................................................................................................
................................................................................................................
................................................................................................................
................................................................................................................
................................................................................................................

ANNEX 2
Sample Content Analysis Codebook

Coverage of the Impeachment Trial of President Joseph Estrada by the Three Leading National Broadsheets

Instructions for Coders
The articles you will be coding include only those that appear in the MAIN SECTION of the newspaper; you will not code articles appearing in the other sections such as entertainment, lifestyle, business, and supplements. Read through the main section and check all the articles that focus on the impeachment case/impeachment trial. Pay particular attention to column articles; each column discusses several topics so you should not look at the headline alone to determine if the impeachment trial/case was tackled in the column.

Count how many articles there are that pertain to the impeachment case/trial. Assign a number to each article.
Before coding an article, scan through it to familiarize yourself with the details of the text. Enter all information in the content analysis form as per the instructions. Use pencil for coding so you will be able to make changes in your entries if needed.

1. Coder’s name – please write your name
2. Newspaper – please write the name of the newspaper where the article appears
3. Date of issue – please put the exact date (mm/dd/yy) the newspaper was published
4. Article number – indicate which article of how many articles in the newspaper issue you are coding, for example, article 1 of 5 articles
5. Type of article – indicate whether it is a:
- news article – factual reports about the case/trial
- column
- editorial
- editorial cartoon
- letter to the editor
- feature article
6. Time started and ended – indicate when you started and ended coding
7. Page location of the article – check whether front, inside or back
8. Type of front page article – indicate whether headline article or not; check NAP (not applicable) if the article you are coding appears in the inside or back pages
9. Location of front page article – check whether above or below the fold; check NAP if the article is not a front page article
10. Length of article – measure in column inches, i.e. the number of columns wide x the number of inches deep. If the article appears on more than one page, add the column inch values per page
11. Impeachment case topics discussed – the choices are indicated on the content analysis form; check “others” if the topic does not correspond to any of the specific topics listed.

If you check “others” do not forget to write the topic is on the space provided.
Remember that an article may tackle more than one main topic. A main topic is one which is central to the content of the article. Topics that are mentioned only in passing in the article should not be checked/listed.

On the box provided (5.1) give the principal details for each specific topic discussed. For example, if the topic is “accusations against Pres. Estrada” the details could be (depending on the content) about what the specific accusations are, profiles of the people linked to the accusation, amount of money involved, etc. Give the principal details for each of the topics you checked in Item 5 of the content analysis form.
12. Impeachment trial topics discussed – similar to the instructions given above, you should check all main topics mentioned in the article, specify the topic if you checked “others”, and give the principal details for each main topic.
13. Personalities/institutions involved – list all principal personalities/ institutions mentioned in the article. Principal personalities/institutions are those who/that are crucial to the discussion/story.

Be as specific as possible in your listing. With the exception of a few personalities like President Estrada, the senators, and the well-known justices, you should indicate the position and agency affiliation of the person listed, or indicate his/her role in the impeachment case/trial.
14. Treatment of the personalities/institutions – treatment is classified as follows:
- favorable – if the overall portrayal of the person/institution is positive, i.e., there are more positive than negative statements about the person or institution
- neutral – if the overall portrayal is factual or “balanced”, i.e., there is an equal number of positive and negative statements about the person or institution
- negative – if the overall portrayal is negative, i.e., there are more negative than positive statements about the person or institution

The information should be entered in the table provided. There should be separate entries for each person/institution listed in Item 7 of the content analysis form.

Please enter the key words that represent the favorable or unfavorable treatment. Remember that if the treatment is neutral, there should be both positive and negative key words in your entries. If you think there are no key words that can represent the neutral treatment, explain why you feel the treatment is neutral.
15. Other comments – we are interested in any other insights you might have about the article you coded. Please feel free to say anything about the topic discussed, the arguments presented by the writer, your personal assessment of the way the issue was covered by the writer, etc.

ANNEX 3
Sample Content Analysis Coding Guide for Data Processing

What is content analysis?
the systematic study of the messages disseminated through the various media

Applications of content analysisin journalism
Compliance with the ethical and legal standards for crime reporting
Compliance with the standards for reporting the results of opinion polls
Adequacy of the coverage of important issues
Adequacy of the coverage of various sectors in society
Portrayal of important public personalities

How can content analysis help the media become more effective watchdogs of society and catalysts for change?

#1
Basis for identifying strengths and weaknesses in coverage of the news and measures that can be taken to improve coverage

#2
Basis for self-understanding

#3
Basis for understanding how we look at our audience and why we look at them that way

#4
Basis for analyzing the probable impact of the media on the public
THANK YOU

Content Analysis Principles and Methodologies

30 November – 3 December 2004
Ulaanbaatar, Mongolia

What is content analysis?
- A research method that uses a set of procedures to make valid inferences from the text (Weber, 1990)
- A research technique for making replicable and valid inferences from the data to their context (Krippendorf, 1980)
- Any technique for making inferences by objectively and systematically identifying characteristics of messages (Holsti, 1969)

Features of content analysis
- A research method
- Focuses on the message or text
- First goal: description identify content and structure of the message evaluate the quality of the message make comparisons across sources or across time
- Second goal: inference character of the source character of the audience effects of the message cultural norms and social behaviors the message reflects

Types of content analysis studies
Classification according to:
- Medium analyzed
- Purpose of analysis content classification content interpretation
- Type of data quantitative qualitative
-Choice of approach depends on objectives of the study and the resources available

Purposes and applications of content analysis
Content analysis can help determine:
- The image of a person, group or institution as portrayed by the media
- The image that a person or institution projects
- How an issue is covered/treated by the media
- Specific issues that are important for the media or public
- Differences in the way different media portray people/institutions or cover an issue
- Changes in media’s portrayal of people/institution or coverage of an issue over time

Steps in conducting content analysis
- Identify topic, problem and objectives
- Identify variables and measures
- Formulate instrument
- Pretest and determine inter-coder agreement
- Select sample
- Gather data
- Process and analyze data
- Interpret findings
- Prepare the report

Identifying the topic, problem, and objectives
Guide questions
What issue will I analyze?
What media will I look at?
What are my research goals?
Why should I do this study?
Do I have the required resources?

EXAMPLE
Issue: Impeachment trial of President Joseph Estrada
Research problem: How did the three leading national broadsheets cover the impeachment trial of President Estrada?
Objectives:
1.To describe how each broadsheet covered the impeachment trial in terms of the following:
- prominence (importance) given to the case
- specific aspects of the impeachment case that were discussed
- specific aspects of the impeachment trial that were discussed
- treatment of the personalities and institutions involved in the impeachment trial
2. To identify similarities and differences in the broadsheets’ coverage of the impeachment trial
3. To present recommendations as to how the broadsheets’ coverage of similar issues may be improved.
Identifying variables and measures
- Refer to the problem and objectives of the study
- List the variables
- Develop operational definitions

Formulating the content analysis instrument
Content analysis form
Content analysis codebook
Pretesting the instrument and assessing inter-coder agreement
Steps:
1.Select an article to be analyzed
2.Ask several coders to code this article independently
3.Compare the coders’ entries for each item/variable
4.Clarify confusions and differing interpretations of coding instructions
5.If desired, compute the inter-coder reliability coefficient

Selecting the sample
Guide questions:
What types of materials should I include?
What time period should I cover?
Will I get all or a sample of the materials?
How will I select the sample?

Processing and analyzing the data
Processing:
1.Formulating the code guide
2.Constructing the data file
3.Encoding the data
Analysis:
1.Generating frequency tables
2.Generating cross-tabulations
3.Computing test statistics

Interpreting the findings
Make the numbers understandable to the readers
Answer the research problem and objectives
Offer possible explanations for the findings
Discuss implications of findings