An If Americans Knew Report Card

Deadly Distortion

Associated Press Coverage of
Israeli and Palestinian Deaths

Preview Edition for American Society of Newspaper Editors
Released Wednesday April 26, 2006

Study Period:
January 1, 2004 - December 31, 2004

Background

In 2003, If Americans Knew¹ began issuing report cards to media across the country on their coverage of the Israeli-Palestinian conflict. We provide these reports to media outlets with the aim of assisting them in covering this topic accurately. In addition, we make the reports public, to help readers evaluate for themselves the reliability of their sources of information on this issue.

This study of the Associated Press Newswire (AP) covers 2004, a year for which we have completed studies of The New York Times, and the major network television channels: ABC, NBC, and CBS.

We chose to study AP because it is one of the major sources of world news for Americans. AP, according to its website, is the world’s oldest and largest news organization. Many newspapers depend on it for their international news.

Method

We recognize that reporting on Israel/Palestine is a controversial topic. Therefore, while there are many possible ways to measure accuracy, we chose criteria that are relevant, conducive to statistical analysis, and immune to subjective interpretation.

We chose to focus on the reporting of deaths, because this allows meaningful statistical analysis that would be impossible in a qualitative study. This unambiguous measure allows us to determine whether AP demonstrates even-handed reporting, regardless of nationality or religious background. Fortunately, accurate data for both populations is available from the widely respected Israeli human rights organization, B’Tselem². We only included Israeli deaths directly caused by the actions of Palestinians, and vice-versa.

This study of AP follows the conventions of our previous studies. Our decision to look at only headlines and first paragraphs was motivated by the goal of assessing the average reader’s experience and the prominence given to the coverage. (In addition, past studies indicate that the patterns found in headline- and lead paragraph-coverage tend to hold when the entire article is examined³.)

We used the LexisNexis database to access all AP articles filed from Israel or the occupied territories (the West Bank and Gaza Strip) during 2004 that went out on the U.S. print wire.

Findings:

I. Coverage of All Deaths

We found a significant correlation between the likelihood of a death receiving coverage and the nationality of the person killed.

In 2004, there were 141 reports in AP headlines or first paragraphs of Israeli deaths. During this time, there had actually been 108 Israelis killed (the discrepancy is due to the fact that a number of Israeli deaths were reported multiple times).

During the same period, 543 Palestinian deaths were reported in headlines or first paragraphs. During this time, 821 Palestinians had actually been killed.⁴

In other words, 131% of Israeli deaths and 66% of Palestinian deaths were reported in AP headlines or first paragraphs.

That is, AP reported prominently on Israeli deaths at a rate 2.0 times greater than Palestinian deaths.

In reality, 7.6 times more Palestinians were killed than Israelis in 2004.

9 Israeli children’s deaths were reported in the headlines or first paragraphs of AP articles on the Israel/Palestine conflict in 2004, when 8 had actually occurred. During the same period only 27 out of 179 Palestinian children’s deaths were reported. (Children are defined by international law as those who are 17 and younger.) Additionally, Palestinian children made up a disproportionately large number of Palestinian deaths in general. Children’s deaths accounted for 21.8% of the Palestinians killed, while children’s deaths accounted for only 7.4% of Israelis killed during this period.

22 times more Palestinian children were killed than Israeli children. AP reported on 113% of Israeli children’s deaths in headlines or first paragraphs, while reporting on only 15% of Palestinian children’s deaths. That is, Israeli children’s deaths were reported at a rate 7.5 times greater than Palestinian children’s deaths.

Comparing running totals for actual deaths and reported deaths once again reveals that while AP’s reporting on Israeli children’s deaths closely tracks the reality, the reporting on Palestinian children’s deaths lags far behind the actual number, following a path similar to Israeli children’s deaths. This is in stark contradiction to the reality, in which Palestinian children were being killed at a rate over 22 times greater than Israeli children.

In order to discover the impact of repetitions on the study, we examined AP’s coverage of children’s deaths without counting repetitions. We found that AP repeated two Israeli children’s deaths once, and one Palestinian child’s death three times. Hence, not counting repetitions, AP covered 88% of Israeli children’s deaths – a rate of coverage 6.5 times greater than their coverage of Palestinian children’s deaths (of which AP covered 13%.)

III. “Clashes” – A Case Study of AP’s Diction

Many qualitative observations may be made about bias in news coverage. One interesting aspect is the terminology used by a news source in reporting on this conflict. We examined AP’s usage of the words “clash” and “clashes”. Of all the conflict deaths AP reported in 2004, 47 deaths were stated to have taken place during a clash. Every one of those 47 was a Palestinian death, which suggests a more unilateral violence than the word is commonly understood to convey.

Additional Notes: Context

While gathering the data for this study, our analysts looked at hundreds of articles that AP published on topics relating to the Israel/Palestine issue, and noted a number of additional patterns that merit further examination. (The daily reports from the International Middle East Media Center, imemc.org, are useful in evaluating AP's coverage.)

1. There appeared to be differentiation in the amount and type of contextual information provided regarding the people killed and the circumstances of their deaths. While Israeli deaths were often depicted as innocent victims of Palestinian aggression, Palestinian deaths seemed more often to be portrayed as a necessary result of conflict.
2. We noticed that several pertinent subject areas had been minimally covered by AP. For example:
• Palestinian prisoners. Torture in Israeli prisons is listed as a concern in the first paragraph of Amnesty International’s report on Israel covering the year 2004.⁵ It was first exposed by the London Times in 1977 and is continually noted by the US State Department, numerous human rights organizations and others.⁶ Over 9,000 Palestinians are currently incarcerated by Israel (over 4,000 have not had a trial),⁷ with the number of Palestinian political prisoners per capita among the highest in the world.⁸ Torture of Americans of Palestinian descent was detailed by Foreign Service Journal in 2002.⁹

Yet, apart from four stories on a prisoner hunger strike, we could find only two stories that described Israeli prison conditions for Palestinians. Only one AP headline from the area mentioned torture – and this one was about Lebanese, not Palestinian, prisoners.

• Israeli Refusers. During 2004 numerous Israelis refused to serve in the Israeli armed forces in the occupied territories.¹⁰ By year’s end there were 1,392 such “refuseniks” and 37 had gone to prison. This movement was a topic of increasing discussion in Israel and the subject of numerous news reports. Yet AP had only one story on this.
• Nonviolence movement. Palestinian resistance efforts have included numerous nonviolent marches and other activities, many joined by international participants, Israeli citizens, and faith-based groups. This nonviolence movement has been an important topic in the Palestinian territories, with growing numbers of people taking part – in 2004 the Palestinian News Network reported on 79 major demonstrations that were exclusively nonviolent. Yet, we did not find any reports in which AP had described a Palestinian demonstration or other activity as nonviolent or utilizing nonviolence.
3. We noticed significant stories that, perplexingly, were sent only on the Worldstream wire, disseminated internationally, but that were not sent to U.S. editors. For example, on May 11, an AP story reported: “The Geneva-based Defense for Children International and Save the Children, based in Sweden, said that as of May 2004, 373 Palestinians under 18 were being held in Israeli detention centers and prisons. At least three of the detainees are under 14...The groups charged that the treatment of Palestinian child prisoners by Israeli authorities amounts to a pattern of violence that has gone unchecked for years...” This story was not sent to U.S. newspapers.

It is unclear to us why this story would be considered newsworthy for readers in other parts of the world but not for readers in the U.S., Israel’s primary ally. A study comparing AP reports sent to U.S. papers to AP reports sent to international papers might be of interest.

Previous studies have shown newspaper coverage often to be significantly more distorted than the pattern we have found for AP,¹¹ and we wonder if AP’s system for alerting newspapers to the top stories of the day may play a role in this differential. We urge newspapers and AP itself to examine this system. We hypothesize that such an investigation would reveal increased distortion.

Conclusions

We are concerned about the results of this study. As the primary newswire, newspapers across the country rely on AP. Since most newspapers cannot afford to send their own correspondents abroad, AP is often one of only a few sources of international news. We believe the readers of these papers, as well as all Americans, are entitled to full and accurate reporting on all issues, including the topic of Israel/Palestine.

Given that AP had ample coverage of this issue (over 700 news stories on deaths alone), it is troubling that so much critical information for American readers was omitted. Further, our findings suggest a pattern of distortion in AP coverage of the Israeli-Palestinian conflict inconsistent with normal journalistic standards. Such a pattern of distortion, in which readers were given the impression that the Israeli death rate was greater than it was, and that the Palestinian death rate was considerably smaller than its reality, may serve to misinform readers rather than inform them.

In particular, our study shows immense distortion in the coverage of children’s deaths. By covering such a large proportion of Israeli children’s deaths in headlines or first paragraphs and such a low proportion of Palestinian children’s deaths, AP’s coverage obfuscated the fact that in actuality over 22 times more Palestinian children were killed than Israeli children.

Now that AP has been alerted to the distortions in its Israel/Palestine coverage, we encourage it to undertake whatever changes are necessary to provide accurate news coverage of this vital issue.

It would be valuable to examine AP’s structure of reporting from the region, its editorial direction from the international desk in New York, and the specific mechanisms AP has in place, if any, to ensure that bias does not intrude on its reporting on this issue.

Finally, in the interest of full and accurate reporting, we urge AP to inform its readers of the findings of this study. In addition, we encourage AP to report the strategies it intends to use in remedying the significant flaws this study has discovered in its coverage of the Israeli-Palestinian conflict.

Summary of Data

End Notes

1. If Americans Knew is dedicated to providing full and accurate information to the American public on topics of importance that are underreported or misreported in the American media. Our primary area of focus at this time is Israel/Palestine. For more information contact us.
2. For more information about this organization, visit their website: www.btselem.org.
3. See our New York Times report for a sub-study of entire articles.
4. 4 These numbers do not include Palestinian civilians who died as a result of inability to reach medical care due to Israeli road closures, curfews, etc. The figure for Palestinian deaths is extremely conservative, since it is difficult for B’Tselem to report on deaths in the Palestinian territories. Palestinian medical organizations report a higher number for this period. For example, the Palestine Red Crescent Society (www.palestinercs.org), internationally respected for its statistical rigor, reports that 881 Palestinians were killed during this time.
5. Amnesty International Report 2004.
6. London Sunday Times, June 19, 1977; The Public Committee Against Torture in Israel; Israel and the occupied territories, Country Reports on Human Rights Practices - 2004, US State Dept; Human Rights Watch briefing to UN Jan 2004; “Prison Tactics A Longtime Dilemma For Israel, Nation Faced Issues Similar to Abu Ghraib,” By Glenn Frankel; Washington Post Foreign Service Wednesday, June 16, 2004, Page A01.
7. According to the Palestinian Central Bureau of Statistics, more than 40,000 Palestinians have been arrested since the start of the September 2000 Al-Aqsa intifada. As of April 2006, 9,400 Palestinians remain jailed in 30 prisons throughout Israel.
8. “The Politics of Prisoners,” Bex Tyrer & Tone Anderson, The Alternative Information Center.
9. “Arab-Americans In Israel: What ‘Special Relationship’?” by Jerri Bird, Foreign Service Journal, June 2002
10. Courage to Refuse, “Combatants Letter”.
11. For example, a six-month study of the San Francisco Chronicle showed a 30:1 differential of Israeli children’s deaths to Palestinian children’s deaths; a six-month study of The Oregonian by AUPHR showed the paper’s headlines had reported Israeli children’s deaths to Palestinians children’s deaths at a rate 44:1.