(upcoming post: Troy Davis Case. International Pressure against execution? Not really)
Troy Davis case
Troy Davis was convicted and sentenced to death, accused of the murder of police officer Mark MacPhail in 1989 in Savannah, Georgia. He was implicated in the murder by other suspected man and arrested four days after the crime. Several eyewitness accused Troy Davis of shooting the police. Two testified that Troy Davis confessed the murder to them. Davis declared always innocent.
Later, some witnesses recanted their previous testimony and described police coercion. After three execution dates stayed. Georgia Board of Pardons and Paroles denied clemency by September 19, 2011. Troy Davis is expected to be executed today September 21, 2011. (update September 22: Troy Davis was executed by lethal injection, 7pm EDT).
Death penalty debate and the media
As it has happened in some precedent cases, doubts about the process have raised a strong movement from people and activists opposed to death penalty. The case is thus creating relevant media attention, not only in Georgia newspapers, but also nationwide in the United States.
Expanded media coverage allows us to analyse this crisis and which kind of information we can derive from the analysis of media behavior.
As in all precedent posts, our unique aim is to provide information and knowledge to people interested in media analysis, reputation impact and crisis management. As for each particular issue treated in each post, we never take any particular position. In this case, we will not present our position in regard to death penalty, as my opinion about it is not relevant for ther readers.
Of course, some readers of this particular post are not mainly interested in media analysis, but about the death penalty debate. Like in all our posts, we try to provide new information that can be useful for understanding the specific topics analyzed.
Death penalty is an open debate in the United States. According to a recent Gallup poll, there is a majority of americans supporting death penalty for convicted murder. According to this source, 64% are in favor of death penalty, while 29% oppose to it. Time evolution shows an increase of opposition to death penalty since mid 90s.
(source: Gallup 2010)
We will show in this post some results from media coverage analysis linked to death penalty debate, using MRI Universidad de Navarra techniques.
Media Attention Index by group of states
We count with information about news published about Troy Davis case in each state in United States. Our strategy is to analyze media coverage behavior by regrouping states by relevant variables to the death penalty debate. This is a strategy that we have also followed in our post concerning New York 9/11 attacks and memorial.
In this analysis we have excluded results from Georgia, as it behaves as a clear outlier as this state is the main source of news and has covered the case in a massive way in comparison with any other state. We have excluded also news from California and New York, as their influence in final results are too big. These two states publish 29% of all news about Troy Davis case.
The basic variable that we analyse is what we call “Media Attention Index”. Media attention index measures how relevant Troy Davis case is in each state. This is measured by the number of news published about the case, in comparison with the media coverage given to a selection of other neutral issues, unrelated with death penalty topic.
If we find similar level of media attention in both groups, this implies that the classification variable that we have chosen is probably no sensitive to the death penalty debate. This absence of sensitiveness is already a result. But it is usually more revealing to find which elements are topic-related, and this result emerges when we find that media coverage diverges in each group of states.
First variable chosen is Republican-Democratic orientation of the states. We could use different measures, as we did in New York 9/11 post. Here we just select the measures that identify as Rep-Dem when there is a margin bigger than 10 points in 2008 presidential election.
Our results indicate that there is a clear partisan media coverage of this death penalty process. Republican oriented states provide a sustained higher media attention than Democratic states.
Now we turn to two other key division of states. First, we look at media attention depending on death penalty status in each state.
We find that states currently with death penalty are following Troy Davis case in with a higher degree of media attention than states without. We can also observe a decrease of the gap, that was unnoticed in with the division Republican-Democrat.
The complementary variable is to regroup states between those with recent executions (year 2010 and 2011) opposed to those without execution, whatever the legal status of death penalty is.
A similar result than before is found out: states with convicted recently executed tend to provide higher levels of media attention to Troy Davis case.
All three results presented show that media attention is sensitive to our variables chosen: Republicans, states with capital punishement and states with recent capital executions tend to show higher degree of media attention to Troy Davis case than others.
As we have shown the time evolution of the selected variables between September 20 and 21, when media coverage explode, we can look inside these results more carefully.
Within the time interval we count with a decisive fact: final clemency denial by Georgia Board of Pardons and Parole. News before this decision refer to the last stages of Troy Davis judicial process and positions of each camp about this case. News after the decision are more focused about the execution itself and its justice. Probably, debate about death penalty is more intense in news after the confirmation of the sentence to death.
We propose to check the results already presented by dividing them into stages before and after denial of clemency.
Results are revealing. First figure shows media attention index by states according the legal status of death penalty. News about the case till the denegation of clemency are more intensely followed in states with death penalty than others. But media attention profile changes if we consider only the flow of news after the decision was announced and execution is fixed for September 21. Then, the states where death penalty does not apply or has been abolished provide a wider media coverage to Troy Davis case than states with death penalty.
If we establish the difference between states with recent lethal executions, we osbserve a similar trend, but less pronounced. Before the clemency decision media coverage was sgubstantialy higher in states with recently executed inmates, while it balances after the decision, but not to the point to reverse the trend.
Update September 22, 2011
As explained before, Georgia Board of Pardons and Paroles denied clemency to Troy Davis sentence to death. Last minute appeals were rejected or non considered. Troy Davis was executed using a lethal injection, by 7pm EDT.
We provide additional results about coverage of Troy Davis case.
First we show the evolution of media attention index during the last hours of the case. As expected in high-profile death penalty case where both sides contend, we observe an explosion of media attention in the last stages of the case. The index dramatically increase from a 50 points value by September 20 to 100 points by September 21 6pm CET. By September 22 10am CET, some 8 hours ahter the execution took place, media attention index reaches almost 200 points.
News about the case are thus concentrated in the period where the execution is confirmed and inminent. Almost 80% of news in US newspapers about Davis case have been published after denial of clemency.
In the last stages, is becomes an all nation issue, and our results show clearly that both camps, in favor and opposed to death penalty cover the case with the same level of media attention. This is in contrast with the initial coverage pattern, that we have shown that it is more intense in sates applying death penalty. This is confirmed in results shown in the following figure, where we find that media coverage is roughly the same in the execution day in both groups of states.
Main components of media storyline of Troy Davis case
In the first section of this post we have used as tool of analysis media attention index. The goal was to identify different patterns based in ideology, legal staus of death penalty and recent executions.
In this and following section we look into the content of news about Troy Davis case. We want to lean which are the key elements of the storyline used by the media.
In this section we show the structure of the main components of media stroyline. In the next session we will analyze to which extent each singular topic suffers from media bias reflecting a sensitiveness to the death penalty debate.
Here we present global media coverage in all states in the United States.
We explore the content of news about Troy Davis case. We select specific topics linked to death penalty debate. We measure the prevalence of each selected topic in newspapers. The higher the value in the figure, the more frequently the specific topic is present in articles covering Davis case.
Main topics are direct references to the execution and to the clemency process and decision, as they appear almost in 9 out of 10 articles.
Direct references to death penalty reaches a value about 60%, which is more than three times more than the use of the term “capital punishment”.
The presence of supporters to Troy Davis case is relevant, as Amnesty International claims is presented in 60% of all articles. References to former FBI director William Sessions, also favorable to not applying death penalty to Davis is aover 30%.
References to personal and social circumstances of the convicted inmate (“racial” or “poor”) are rather marginal in the storyline as they take value 7.7% and 2.9% respectively.
Finally, the critics about the outcome of the case, considering that Davis “is innocent” or that an injustice is taken are present in some 20% of all articles.
Troy Davis case storyline by death penalty status of states
Now we move to a complementary tool. We measure the prevalence of each selected topic in newspapers comparing states with death penalty against states without death penalty.
Our aim is to identify the sensitive issues in the death penalty debate. If prevalence is different between groups, this media bias indicates that this specific issue is relevant for the death penalty debate.
We have included here results about coverage by media in Georgia. Of course, Georgia pertains to the groups where death penalty is legal and should behave like this group of states. But in the same time, as Davis case is a local case, debate and controversies arise in a clear wide extent, and makes less predictible its media profile.
It is important to take into account when considering Georgia results in comparison to behavoir of other states that this is an outlier case and has its own dynamics. Besides the increase of the controversial profile of the news already mentionned, the richess of the debate and the coverage of the case in Georgia produces as a result that prevalence to single topics will tend to be lower in general than all other states. This is due to the fact that other states do not cover all the issues treated in local media and concentrate in the core topics of the case and the debate. This downward bias requires to be always in mind when analysing Georgia results.
If not noticed otherwise, empirical results refer to news up to September 21, before the execution took place.
Moving to the results, we present first results concerning the core issue of Troy Davis case: it is a death penalty case.
We find that direct references to death penalty are proportionally higher in states with death penalty. In Georgia the rate is lower. As explained, as newspapers in Georgia are covering the case well before the last days, they include news not directly talking about the sentence.
The reference to “capital punishment” instead of death penalty behaves in the opposite way, as it is more profusely used by states without death penalty. Using the capital punishment term is somehow euphemistic as it allows to avoid the mention to death. In contrast it could produce stronger fellings in readers, as it refers to “punishment” instead of “penalty”. Both terms are evidently related, but focus in different aspects.
Direct reference to the execution is again more used in coverage of states with death penalty.
Like in all constested convicted cases, last instance of clemency decision plays a major role in coverage of death penalty case. Our results show that references to clemency is not a partisan element of the storyline, as it is used in the same level in both sides.
When doubts emerge about the fairness and certitude of a death penalty sentence, there is media space for people and movements against death penalty or just supporting the convicted inmate. It is interesting to measure the media prevalence of supporters of Troy Davis interests.
We measure the references to Amnesty International, and to former FBI director, William Sessions. Both have had a prominent and dirct implication calling for not applying death sentence to Davis.
We get a result that is opposed to our own expectations and is counterinutitive. Media from states with death penalty provide higher coverage to Amnesty International than states without. Surprisingly for us, more space is given in states where AI is considered an opponent than in states where it is viewed as a case supporter. Those are results before execution and gap is not substantial, but they are really intriguing to us.
References to William Sessions’ position is higher in states without deatn penalty. We also see a drop of media interest in Georgia.
Next element is the prevalence of issues that are usually present in the death penalty deabte in the United States: the racial and the poor condition components. They are both relevant for Troy Davis case, as he is a black man from a poor area.
As expected, both elements suffer from media bias and clearly indicate that they are issues that are more present and used by states opposed to death penalty than states with death penalty.
The final issue we want to chek is the global assessment of the case. In the following figures we present the prevalence of direct references to “is innocent” and to “injustice”. Those are terms that are advanced by people and institutions opposed to Davis’ execution and critizise the fairness of the legal process.
“Injustice” is mentioned more times in newspapers from states without death penalty, as expected.
But, surprisingly for us, we find again that mentions to “is innocent” are more common in coverage by states with death penalty. Look also to the behavior in Georgia, where this issue is disproportionately highly present.