Blog Functions as Risk and Crisis Communication During Hurricane Katrina


  • The authors would like to acknowledge and thank the University of Georgia graduate (JongSuk Choi and Hana Kim) and undergraduate (AnnElise Cutrer, Anna Felz, Leah Kapa, Victoria Salegna and Meghan Wright) students who spent countless hours amassing and coding the data set for this study.


Blogs were examined during the 2 weeks after Hurricane Katrina hit the U.S. city of New Orleans to better understand what risk and crisis communication functions they served. The 4 major functions—communication, political, information, and helping—included both filtering and linking about rescue needs and efforts, missing persons, ways to offer and find assistance, fostering community, and providing information on damage and government response. A thinker function was fulfilled where bloggers expressed opinions, especially on government response. An additional function not previously delineated was identified, which might be termed emotive or therapeutic. In addition, several of the blog functions indicate the role the Internet plays in maintaining a sense of community in times of crisis.

The term “blog,” short for “Weblog,” was coined in 1997 to describe an online journal with dated entries and a mix of links, commentary and usually, reader responses (Blood, 2000). In the decade since, blogging's explosive pop larity has been well-documented. One website that tracks current trends in blogs,, estimates there are now more than 100 million blogs in existence, with more than 50,000 created every 24 hours (“Blogpulse Stats,” 2009).

By their sheer numbers alone, blogs demand the attention of communication researchers. Until now, however, most research has narrowly focused on the sometimes-adversarial relationship between blogs and the traditional media, or on the power of blogs to reveal scandals or corruption that have not been widely reported elsewhere. But there are a number of blog functions that are not yet well understood, including their potential role as information conduits and social networks during crises and emergencies when traditional information channels and communication devices may not be available.

In an effort to better understand blogging in an emergency and risk communication context, this study examines blog use over a 2-week period during and after Hurricane Katrina, the August 2005 Category 4 storm that killed more than 1,500 people (Martel, 2006), displaced about 80% of New Orleans' population (Vanderford, Nastoff, Telfer, & Bonzo, 2007), and caused at least $100 billion (McMillan, 2006) damage to the U.S. Gulf Coast. The primary goal of this study and its unique contribution to the literature is to develop a typology of functions for blogs as crisis communications. This area of research is still in its infancy and would benefit from a descriptive framework for understanding how blogs are used in crises.

Functions of Blogs: Attempts at Typology

An ethnographic study observing people in their homes found that individuals have both “a social as well as an informational relationship” with the Internet (Carey, 2005). The Web has wended its way into every part of the day, and for many users has become as much a part of their routine as brushing their teeth. Particularly for users with irregular schedules and little free time, the ability to go directly to content they choose, dipping in and out in slices of time they determine, has helped the Web displace other forms of media usage. Compared to other types of media, relatively little is known about Web usage and even less is yet known about blog usage. As a new medium to be analyzed and understood, computer-mediated communication has revived the utility of uses and gratifications as a theoretical model (Kaye & Johnson, 2002; Ruggiero, 2000; Tewksbury & Althaus, 2000), therefore making it an appropriate lens through which to view the functions of blogs.

Scholars have identified numerous uses of the Internet that are similar to traditional media: diversions of all types, including relaxation, entertainment and escapism, information and social utility, as well as uses that may go beyond those of the traditional media, such as interactivity and interpersonal connectivity (Eighmey & McCord, 1998; Ferguson & Perse, 2000). Song, et. al. (2004) adds to this list of gratifications with “virtual community,” in which users “use the Internet to form new relationships that are totally different than their real-life relationships” (p. 391).

Uses and gratifications of the Web do not necessarily align in the same ways for differing socioeconomic groups (Jaeho, et. al., 2003). Among higher socioeconomic groups, some of the more common uses such as interaction, surveillance, and consumption appear to gratify needs for connection, learning, and acquisition, respectively. However, lower socioeconomic groups may tend to use interaction as a means of learning–perhaps, for example, gaining information from chat rooms, bulletin boards, or blogs—and consumption as a means of connection–such as perhaps gaining social gratification from using E-bay (Jaeho et al., 2003).

Like other media, use of the Internet has been shown to increase in the aftermath of a disaster (Aikat, 2005; Boyle et al., 2004). In their study of uses and gratifications applied to consumption of a specific news event, Boyle, et al. (2004) suggest that uncertainty reduction is one of the key reasons for information-seeking. Such media use is positively associated with the degree of emotional reaction to an event but the causal relationship is unknown; a greater initial emotional response may result in more information-seeking, or many hours or days of attending to news coverage may lead people to respond more emotionally (Boyle et al., 2004; Riffe & Stovall, 1989).

Like other Internet content, blogs appear to serve a multiplicity of functions for users, from information to entertainment. Although blogs are a subset of Internet content, their diversity shows that in many ways, they represent a microcosm of the functions of the Internet itself.

No single taxonomy describing blogs has yet been officially adopted as the standard, but a number of people have attempted categorization. In their simplest form, blogs have been widely defined as either thinker (original information and opinion) or linker (social and informational networking) or, alternatively, by some as journals–that is, pontifications of varying lengths that are updated regularly, often multiple times a day—and filters, “a mechanism for selecting, evaluating and aggregating information across the web” (“Caslon Analytics,” 2006). Some scholars have delineated blog types in more detail: Iverson (2006) identified five major functions: filter blogs, which narrow down a significant body of information; knowledge blogs, consisting of factual, often searchable data; news blogs, with frequent updating; meta blogs that summarize a number of other blogs; and social networking blogs, such as Hargittai (2003) also divides blogs into five types: the personal journal, links galore, interactive commentary, one-way commentary, and hodgepodge, which includes elements of all types.

Still others have added to the list of functions with labels such as journalist, citizen journalist, corporate, political, PR/advertising, professional, confessional, conversational, muse, and community forum (Comerford, 2004; Fernando, 2004; Nardi, Schiano, Gumbrecht, & Swartz, 2004; “The Webby Awards: Category Definitions,” 2006), among others. Hartelius (2005) emphasized the community-building function of blogs, proposing a taxonomy based on content, rather than structure, with a continuum extending from topic-oriented to personally oriented. She distinguishes between group blogs, which she likens to bulletin boards or discussion groups, and personal blogs—either supplementary, extending a person's offline life such as vocation, hobbies or causes; or individualizing, the type of deeply personal and detailed opus she terms the “Here-Is-What-I-Had-For-Lunch” blog (Hartelius, 2005, p. 81). Given the variety of categories and the lack of consensus at this time, situational-grounded typologies may be a more appropriate way to sort blogs by function, especially when considering the unique characteristics of a disaster.

Blogs and Disasters

There have been few major studies thus far of the use of blogs in disaster situations, but blogs have begun to make an impact. One of the first traditional journalists to blog during a major breaking news story coincidentally did so during a major hurricane (Singer, 2005). During Hurricane Bonnie in 1998, Jonathan Dube, then of The Charlotte Observer, provided updates to his newsroom approximately every half-hour, providing news and information that was not only more current but also more extensive than what might end up in the next day's print edition (Singer, 2005). In the unique environment of a hurricane, the constant updates also had a practical function, ensuring that as much news as possible was reported before telephone and electrical connections were cut off by the storm (Scanell, 2003).

PR practitioners have also long been aware of the potential uses of blogs to reach the public during emergencies, even if they have not had many opportunities thus far to put the idea into practice (Marken, 2005; Rubel, 2005; Smudde, 2005). For publicists and publics alike, blogs can be particularly effective in sidestepping technological, economic, and political obstacles to diffuse information in a crisis. Many of the first-person accounts of the Asian Tsunami came from bloggers (Ramos & Piper, 2005; Thelwall & Stuart, 2007), not only sharing stories of survival and aggregating news coverage, but enabling rescue efforts and humanitarian assistance, and facilitating the grieving process. Ramos & Piper's index of tsunami-related blogs provides URLs and brief descriptions to database professionals, emphasizing the increase in blog activity connected to the disaster, but does not seek to analyze the content. Thelwall & Stuart's automated analysis tracked communication-related technology in blogs and other media during the London bombing, Hurricane Katrina, and the Pakistan earthquake and primarily looked at what communication technologies were discussed in the blogs about these crises (Thelwall & Stuart, 2007). They suggested that future research methods examining blogs should include qualitative content analysis.

Private organizations also utilized blogs in the wake of Hurricane Katrina: State Farm Insurance, for the first time ever, made use of an employee-only blog to help track missing associates and promote the exchange of information within its Southern Zone, which included the hurricane-ravaged areas (Pryor, 2006).

Blogs are well matched to emergency situations in which other forms of communication fail because the technologies that enable blogs, including the Internet and text-messaging, tend to be more resilient. When land lines, cell phones, and traditional utilities failed in New York City on September 11, the text-messaging capability of the BlackBerry proved to be one of the most effective sources of information exchange between individuals (Krim, 2001; Romero, 2001; Tam, 2001). Because uploading and downloading of Web content can occur wirelessly on the BlackBerry, it is also an effective vehicle for mass communication, including the dissemination of information through blogs. During prolonged power outages, the BlackBerry can run 24 hours a day for up to a week off a single charge (Holsendolph, 2001).

Disaster sociologists and emergency risk communicators define disaster not only in terms of its physical impact–property damage and bodily injury or death–but also in terms of societal disruption (Kreps, 1984; Fritz, 1961). While media play an obvious informational and amplifying role in disasters (Kasperson et al., 1988), examination of emergency risk situations in a sociological context also reveals the media's important latent role: providing emotional support, companionship, and a sense of community for individuals and/or family groups isolated by the crisis (Perez-Lugo, 2004). Researchers have frequently noted that emergent groups may supersede the importance of established societal groups in a crisis; increasingly, these emergent groups may coalesce in social media such as blogs (Kreps, 1984). However, how blogs function in crisis is still poorly understood by emergency planners and risk communicators.

Herring et al. (2004) used both quantitative and qualitative methods to examine the interconnectivity of a broad universe of more than 5,000 blogs used during Hurricane Katrina, and found that while certain subsets did show the characteristics of true community dialogue and connection, the vast majority were “social isolates” (p. 10). However, their survey data do not fully represent the extent to which blogs were used as only 3% of their New Orleans sample reported accessing blogs during and after Katrina. While 3% seems like a small number, we do not know what the noncrisis baseline is to compare it to. Also, that 3% may have been posting multiple messages a day or even an hour over the course of several days/weeks.

No studies have specifically examined community-building functions of blogs as a medium during a crisis or disaster, although one study did examine the role the Internet played in maintaining a sense of community, communicating when traditional means are compromised and expressing emotions during a crisis, specifically Hurricane Katrina (Procopio & Procopio, 2007). Given that blog technology may provide an important “redundant” or backup means of communication in a disaster situation and that blog usage appears to increase significantly during crises (Thelwell & Stuart, 2007) (both number of individuals and postings per individual), it is critical for health and risk communicators to develop a better understanding of the uses and characteristics of blogs. People who would normally show no interest in blogging may turn to it in times of crisis when other communication channels are unavailable. It would also be helpful to better understand who the individual bloggers are and where they lived. As discussed in the literature review, the functions of blogs in general have been a research topic that needs to be extended to crisis communication blogs so as to better understand their purpose. Although another type of function, emotions are examined in a separate research question because of their multifaceted nature. Emotions are an integral part of crises and social media like blogs may serve an important therapeutic role. Thus, this study sets out to explore the following research questions:

RQ1: What are the characteristics of the individuals who posted blog entries about Hurricane Katrina?

RQ2: What were the functions of the Hurricane Katrina blog posts?

RQ3: How prevalent was the expression of emotions in the blog posts? What emotions were expressed?


Sampling Procedure

We used a multistage sampling procedure for this study. We first developed a list of all identifiable blogs related to Hurricane Katrina by thoroughly searching the Internet using various search engines (Google, Yahoo!, metacrawler, etc.) and a variety of keywords (e.g., Hurricane Katrina, hurricane, New Orleans, Louisiana, gulf coast, etc.). Some of these were blogs created solely for Katrina and others were existing blogs that began focusing on Katrina after the hurricane hit. In addition, we tried to locate both blogs posted by official media sources (e.g., WWL-TV, a television station and news department in New Orleans) as well as unofficial, personal blogs (e.g., LiveJournal, Blogspot, Yahoo, etc.). Our conceptualization of blogs was broad for several reasons. First, 2005 was still early in the history of blogs and they are more narrowly defined and used today. Second, the utility of message boards and blogs became virtually identical during this crisis situation—a vehicle to communicate a message.

Due to the extreme volume of blogs (over 30) and posts (thousands), we narrowed the focus to only blogs focusing on New Orleans and only the 2 weeks after the hurricane made landfall (8/29/05 to 9/12/05). These choices were made because New Orleans was affected the most compared to surrounding areas and the first 2 weeks should represent the communications related directly to the initial crisis (Hurricane Katrina). In addition, the number of posts went down dramatically (often 80–90%) after these first 2 weeks and the postings became redundant conceptually. We included all three official media blogs and employed a systematic probability sampling procedure to choose one-third of the unofficial blogs (i.e., every third blog was chosen from the randomly selected start). We felt it was important to include all three official media sources since there were so few, they were different conceptually from unofficial blogs and this “official” perspective might be important to fully and accurately capture. The remaining blogs were sampled using probability methods to representatively reflect the population of New Orleans Katrina blogs. Finally, we also systematically chose one-third of all the posts in each of those blogs. Therefore, all days and all blogs were sampled equally. We feel our sampling procedures represented the range and depth of content without being overly redundant. The sampled blog posts were saved in word documents. The final sample included 2,450 individual posts (unit of analysis).

Sample Description

Eleven blogs were included in the sample. Three were official media sources (WWL-TV, WWOZ FM radio and The Times Picayune) and eight were unofficial sources (e.g.,,,,,,, and Of the 2-week period we included, the mean number of posts per day was 163, ranging from 53 on September 10 and 12 and 364 on September 1. The highest number of posts was during the week after Katrina made landfall. Table 1 includes the number of posts on each day and major events related to the hurricane.

Table 1.  Timeline of Sampled Blog Posts
DateNumber of postsMajor New Orleans Katrina Events (AP, 2005)
8/29/200571• 7am cdt—Katrina makes landfall as a category 4 hurricane
  • Katrina rips two holes in Superdome roof
  • Late morning—one levee breached
8/30/2005177• Mass looting reported, security shortage cited
  • Second levee breaks flooding 80% of the city
  • 50,000 to 100,000 believed stranded in New Orleans. Louisiansa Gov. says all remaining must be evacuated.
8/31/2005342• Tens of thousands trapped in Superdome; conditions deteriorate
  • 3,000 stranded at convention center without food or water
  • First group of evacuees at the Superdome taken by bus to Houston Astrodome.
  • New Orleans police call off search-and-rescue to combat looting
  • Health and Human Services secretary declares federal health emergency throughout Gulf Coast, sends medical supplies, workers
  • Pentagon mounts search and rescue, sending four Navy ships with emergency supplies.
9/1/2005364• Looting and violence spread
  • Military decides to increase National Guard deployment to 30,000.
  • Outside convention center, sidewalks are packed with people without food, water or medical care.
  • New Orleans mayor issues “desperate SOS” for more buses to evacuate.
  • Hoping to leave with the bus evacuation, crowds at the Superdome swell to 30,000 with an additional 25,000 at the convention center.
  • The Houston Astrodome refuses to take any more refugees after accepting 11,000.
9/2/2005315• Thousands of National Guardsmen arrive in New Orleans in truck convoys carrying food, water and weapons.
  • Congress approves first $10.5 billion for immediate rescue and relief efforts.
9/3/2005148• Bush orders more than 7,000 active-duty military forces to the Gulf Coast.
9/4/2005145• Evacuees in Texas number more than 230,000. Texas Gov. Rick Perry warns his state is running out of room.
9/9/200587• Amid widespread criticism of the agency's response to the hurricane, Homeland Security Secretary recalls FEMA Director Michael Brown to Washington. Coast Guard Vice Adm. Thad Allen replaces Brown as head of the hurricane relief efforts.
9/12/200553• Brown in Washington announces his resignation from FEMA.
  • Officials report 515 deaths from Hurricane Katrina so far, including more than 40 bodies found in a New Orleans hospital.

The times that the blog entries were posted ranged throughout the day—only 10% of the posts were before 6 a.m., 24% were between 6 a.m. and noon, 36% were between noon and 6 p.m., and 29% were between 6 p.m. and midnight. The number of words in the blog posts ranged greatly between one word and 3,767 words with a mean of 103 and median of 57 words. Seventy percent of the posts had fewer than 100 words and 22% had between 101 and 250 words. The vast majority of the posts used only text to convey their message. Only 2% (44 posts) used pictures and less than 1% used either audio (7 posts) or video (14 posts).

Code Sheet Development

A code sheet was developed for the content analysis using a primarily inductive process. Several of the variables were quite straightforward and necessary to describe the sample, including characteristics of the post (number of words, inclusion of multimedia, etc.) and characteristics of the blogger (experience level, location, gender, etc.). The “emotions expressed” were coded using a previously developed coding scheme by Macias, Lewis, and Smith (2005), which reported a content analysis of health message board discussion threads, as a starting point. Several additional emotions were added based on an informal review of the blog content (e.g., grateful, hopeful, etc.) because of the differences in topics being discussed—health concerns versus survival in a natural disaster.

Finally, the codes for “functions” were developed inductively by reviewing the content of blog posts not included in our sample but that fit the study parameters. There were 22 individual functions that were coded, including calls for rescue/help, offering rescue help, comment on government response, attempts to foster community, communicating with family/friends through blog. Each function is listed in Table 2 and discussed individually in the results section. Detailed definitions were developed (available upon request). The 22 functions were inductively grouped into four main blog crisis communication uses (communication, political, information, and helping) during the writing and editing phases of the manuscript. In other words, we examined all 22 functions for commonalities and identified a natural grouping into the four broader uses. This helps make the research more conceptually digestible. While the communication and information functions are particularly similar to those discussed in previous literature (including uses and gratifications), the process we used was very inductive. We let our data drive the results. The relation of our four functions and those found in previous literature will be discussed further in the discussion.

Table 2.  Functions of the Blog Post
Looking for missing45918.7
Communicating through blog642.6
Call for rescue522.1
Documenting “life experience” related to Katrina391.6
Comment on government response32713.3
Comment on looting1686.9
Information about government response1154.7
Question about looting12.5
Posting official news2369.6
Providing information about damage1777.2
Checking on status of loved one's area1064.3
Providing information about missing that had been found582.4
Asking about damage, no tie to area522.1
Checking status of one's home441.8
Providing information to help locate missing331.3
Providing information about how to get help241.0
Providing social support1968.0
Offering assistance1064.3
Organization of assistance963.9
Attempts to foster community753.1
Organization of rescue help301.2
Offer rescue help8.3

Procedure and Reliability

The coding was done using five undergraduate students. They were trained on the code sheet, definitions, and procedure. To clarify any questions and pretest the code sheet, coders were then asked to code a pretest sample (not part of the main sample). The pretest yielded few questions and high reliability (94% using percent agreement among all five coders). Confusions or problems were resolved.

Due to the volume of the sample and to reduce paper waste, the human coding was recorded electronically. The code sheet was a spreadsheet file with each column being a variable and each row a blog post. The posts were coded directly from Word documents of the blog posts cut and pasted from the Internet. In addition to recording each code on the spreadsheet, the coders also recorded each “function” code in the Word document by selecting the text representing each code and inserting a “word comment.” This facilitated the authors locating instances of each code for analysis. The coder could decide a single post could serve more than one function and would be coded as such.

Approximately 20% of the sample was coded by two coders (all possible pairs of coders was used) in order to establish reliability. Intercoder reliability was established using percentage agreement. Percentage agreement ranged from 51% (e.g., blogger experience level) to 100% (e.g., presence of pictures or information to help) with an overall average of 91%. The majority of the variable averages were above the minimal agreement levels of 80%. Only one variable was below the 70% cutoff and was dropped from the analysis (blogger experience level). Disagreements among the coders were resolved through discussions until 100% agreement was achieved.


Characteristics of Blogger

Fifty-six percent of the posts were from the owner/originator of the blog and 44% were a response to the owner. Most of the bloggers were not New Orleans residents. The location of the blogger could not be determined in 64% of the posts. Of those that could be coded, about 1% were from New Orleans and stayed during the hurricane (31 posts) and 4% (92 posts) were New Orleans evacuees. Those who stayed in New Orleans during Katrina used various means to get their blog messages posted, including generators, a computer with a battery and a fax routed through Texas. Fifteen percent of the bloggers had no familial connection to New Orleans and 13% had family from New Orleans. Over 2% of the posts were made by former or part-time New Orleans residents. Out of the 18% of posts that were made by individuals outside New Orleans, 28% were from the Gulf Coast Region, 59% lived somewhere else in the U.S., and almost 12% were outside the U.S. The gender of the blogger was identifiable about 46% of the time (43% were male and 57% were female).

Functions of Blog Posts

Table 2 lists the frequency which each coded function was present. Examples of individual functions are then presented to give the reader an idea of what these posts included. Table 2 and the discussion of posts is organized into four main functions—communication, political, information, and helping—and then in order from the most frequent to least frequent. The purpose of the four functions is to help make sense of the use of blogs during a crisis. While the original 22 codes were mutually exclusive, the four broader functions are fluid and are not intended to be mutually exclusive. Thus, they are better represented as a Venn diagram as it shows overlapping, interrelated elements (see Figure 1). Therefore, representative quotes from one category may seem slightly repetitive from previous categories in an effort to keep the context of the individual quotes.

Figure 1.

Crisis Communication Blog Functions

We left the quotes as they were originally typed, including typographical errors and all capital letters in order to retain the original intent and tone. However to protect the individuals identities, we removed all personal identifiers and replaced them with “Name,”“Street Address,”“xxx-xxx-xxxx” (for phone, cell, or fax) and/or “e-mail.” We would like to point out that the level of personal detail people were willing to post was incredible. There was often little concern for anonymity in the desire to find loved ones or offer help.


The need to communicate using alternative means (e.g., texting, blogs, etc.) was really the driving motivation for this study. Early reports showed individuals stranded on rooftops in New Orleans after Katrina trying to communicate via cell phones. While this is similar to the interpersonal connectivity and uncertainty reduction purposes found in previous Internet studies (Eighmey & McCord, 1998; Ferguson & Perse, 2000), the situation and context influences the communication purpose and makes it somewhat different.

Looking for missing These posts were plentiful shortly after Katrina hit New Orleans and mostly consist of brief listings of a person's name, any pertinent information to help locate them and a request to be contacted if someone had information on his/her whereabouts. Typically these posts are addressed to anyone with information:

  • I'm looking for my friend, Name, from New Orleans. Does anyone know him?
  • Name worked at Napoleon House—if you know him or have seen him, could you please respond? Thank you.
  • If anyone should know the where abouts of Mr. Name, approx. 6’, light complexion african american male, age 51. Last address known to me is Street Address New Orleans, La. please contact me at Email@HOTMAIL.COM OR Email@YAHOO.COM. he is a very special friend and I just need to know that he is alright.
  • my aunt is on the 7th floor of the doubletree –any info on that hotel? Thank you!
  • Story: My Aunt Name, who lives on Street Address, has not been heard from since Monday morning. She chose to stay home because she wouldn't desert her pets, and we are very worried about her. Would anyone with any info on her whereabouts please contact me at: xxx-xxx-xxxx, or xxx-xxx-xxxx.
  • Does anyone know where the patients from the Southeast Louisiana Hospital, in Mandeville, have been relocated to?

Communicating through blog A number of people utilized blogs in an attempt to communicate directly with friends and loved ones. It was unclear from the data whether bloggers had reason to believe the intended recipient was likely to be reading that particular blog or whether they were simply utilizing every means of communication available to reach someone:

  • DAD PLEASE CONTACT US; xxx-xxx-xxxx OR xxx-xxx-xxxx YOUR KIDS
  • Hi! This is your cousin Name in FL, hoping your family is well. Your other cousins, here in FL, are concerned, too. Please let us know. Your Uncle Name & Aunt Name are my parents-they send their love, as we all do. Let us know if we can help.
  • this is Name in phoenix, if you are ok let me know, xxx-xxx-xxxx
  • Name is alive and well in El Paso. Contact me via e-mail or at my home phone (xxx) xxx-xxxx and I will help you reconnect.

Call for rescue Many of the posts that appeared in the aftermath of the hurricane on the Weblog of the New Orleans Times-Picayune,, were closely monitored by rescuers and are credited with saving lives (Glaser, September 13, 2005). Some posts included specific information about locations and conditions of individuals in need of help, and indicated recent contact with the people in peril:

  • I received a brief cell phone call at 7:30AM this morning from my daughter, who is trapped and waiting to be rescued, along with several other peolpe by the lake behind the UNO campus. Someone please send HELP!!! My daughter's name is Name.
  • 12:00–EXACT ADDRESS OF PERSON STRANDED AT WASHINGTON AND TOLEDANO: Street Address, Zip code. family is trapped on roof
  • As of 9/1 at 8:00 a.m., the Name family- including an elderly couple and three babies- are stranded on a rooftop at St. Roch and Claiborne. Please send help.

Several posts came from trapped individuals themselves and indicated that fear of crime was a significant factor in their plight:


Pets were also mentioned as an obstacle to rescue; some had apparently turned down earlier attempts to be evacuated because their pets were not included. Several posts begged for assistance to rescue pets left alone. Other “calls for rescue” seemed to reflect desperate attempts to confirm the whereabouts and safety of people who were missing but were not necessarily confirmed to be trapped:

  • Please help send rescue to the Locket School on the corner of Piety and Louisa street. My Grandmother–Name and others may be on the 2nd floor of this school. They may also be at Street Address New Orleans Zip code.
  • Name“Gonzie”Name and her elderly sister may still be in their attic at Street Address in Parish. No word from them in over a week. Pls check on them. Their address was left with Plaquemines Sheriff before the storm. Household includes many cats. Pls contact me with any word. Many friends across country are eager for news….

Many of the posts included personal background information, serving to humanize the trapped people. Medical and economic hardships also figured prominently:

  • … My brother has MS and is in a wheel chair. I was no able to speak with my mom for long but she said he was not doing well. … My mom has no car or phone. My brother needs medical assistance. I'm in NY and have no one in Metaire to help me.
  • … My father is disabled and 65 yrs old he has long black hair, ususally in a pony tail, missing front teeth he is a diabetic, has high blood pressure and other health issues….
  • … Name1 is insulin dependent and is not well, Name2 has heart condition, daughter five months pregnant and has gone into premature labor. Please send rescue to the harvey area, they called from payphone, they cannot get out, they have no gas for their water, no food and little meds to last….

Documenting “life experience” related to Katrina Posts sometimes used personal anecdotes to drive home points to their readers. One blogger, for example, talked about preparing for tornadoes in his native Oklahoma:

If you hear a tornado siren blare, by golly you move…. You know to bail out of your car (if you’re out in a tornado) and lay flat in a ditch…. They practice tornados drills in schools and in all major buildings on a regular basis…. We have had massive F5 tornados that cause massive destruction with hardly any deaths. The tornados plant cars on rooftops and wipe out entire towns. Why don't more people die? We’re ready.

One survivor related how his life had changed since the hurricane:

I was sitting in my house on my favorite chair, in front of my computer, eating food from my refrigerator, on my phone, typing my lesson plans for my Math class for the upcoming week. Now I have none of those luxeries: No home, no chair, no computer, no phone, no job, no students to prepare lessons for…


While related to the next function and a typical uses and gratifications function (information), political themes emerged as a distinct category given the government, including law enforcement, involvement in this crisis.

Comment on government response Perhaps no category engendered more emotion than the “comment on government response.” Government at all levels was castigated and government at all levels was defended. In addition to commenting on the specifics of the government's response to the disaster, some posts used the issue as a launching pad for views on other subjects, from the war in Iraq to attitudes toward the media or beliefs about racial or economic differences:

  • … More should have been done by the govt, alot more and alot quicker. I also NOW dont have much confidence in the US, and the presidents response in my opinion has been weak, weak, weak. I hope and pray that my family & my pets are never in a situation like the folks in NO find themselves right now, for it's clear we would be on our own. A rude awakening to us all.
  • … It has been so painful watching this unfold on the news—both because of the human suffering reflected, and also because of the painfully slow and apparently incompetent emergency response leadership at all levels of government. Anybody in charge?
  • Where is the federal aid, where is the food & water, where is Waldo (George W)???? They keep talking about sending in troops and restoring order, but they aren't sening in food and water and supplies. They can drop ship supplies into Indonesia and Beiruit but not in our own backyard?? AND they stopped the buses going to Texas, AND they haven't even begun to sandbag the levee.
  • … I wonder how much cronyism and corruption lead to the complete lack of local and state support for the hurricane. People are blaming the federal government but I am really ashamed of the local people….
  • Impeach Ray Nagin now for gross incompetence. The mayor, ordered a “mandatory” evacuation a day late, but kept the city's 2,000 school buses parked and locked in neat rows when there was still time to take the refugees to higher ground. The bright-yellow buses sit ruined now in four feet of dirty water.
  • … Then I read Ray Nagin's WWL transcript. Finally! No bullshit, so spin. Just real. He spoke from the heart and revealed a genuine concern for those suffering in New Orleans. I had never perceived that kind of genuine care for people from any other politician. For the first time, I felt that the stranded had a voice, someone courageous enough to stand up for them when no one else in power really seemed to care about them….
  • The Iraq War costs the U.S. $186 million per day—to kill people. Can't we allocate enough money, Bush & Co., to save the lives of our fellow Americans in New Orleans?
  • I would rather spend 100 billion on re-building New Orleans than 100 Trillion on a defunct Social Security System that robs the working class and gives pathetic return on hard work.
  • Katrina, as bad as it is, should be a wake up call for this country. Here is a prime example of what you get when you build a help wagon and load it with program after program that stagnates people, black and white, These people for generations have waited/expected the federal government to take care of them. This time the job was left to an enept mayor and city officals. Fire the whole lott… When we start rebuilding the city let there be jobs for citizens in NewOrleans and not big political contractors from washington dc.

Comment on looting These posts tended to come later and were mostly written by people not directly affected by the hurricane. Often they were part of longer posts commenting on several aspects of the hurricane and fallout. Posts were on both sides of this issue and tended to be heated and emotional:

  • And what about the looters and others who were causing more damage? Looting for food is one thing, but looting for TVs and other items was just stupid. How many of those TVs were allowed on the evacuation buses and planes? And without electricity, how were the looters going to watch TV? Those who fired upon people coming to help them deserved to be shot themselves–and some were.
  • Yes I can understand looting in the face of a disaster of this magnatude, if we are talking food, bottled water, baby needs…… but a bucket o beer….tv's…. real. Haily Baber has the right answer to this….martial law,,,,and if need be shoot the looters. These up standing citizens when relocated to other areas and states will continue being such up standing moral citzens. Don't tell me they do not know better. They know right from wrong. And no matter the spin you put on it, the are vermin, each and every one of them. They hamper efforts of those who are trying to help.
  • … as for the looters, it matters not their color but that they need to stop or be stopped.
  • … The hospital was running out of food. Our neighbors were running out of diapers. Life was difficult. Okay. Shit goes awry. Me and Name looted food and toiletries.

Information about government response These posts were primarily passing along information about what the government was doing to respond to the disaster, as opposed to the posts which commented only on the value of the government's response. They were often posted by people immediately affected by the disaster and often provide a first-hand account because they were still in New Orleans.

  • Just saw 2 CH-53E Super Stallion Helicopters pass by overhead, and now on cam you can see what looks like a whole batallion of troops roll toward the Convention Center. . . This convoy coming down the street is loaded with supplies. I see MREs and water and I assume ice. Ok, so the troops used to restore order went in first and now the supplies are coming for orderly distribution (I hope). Hope is on the way for the people at the Convention Center. Finally.
  • The USS Comfort, the Navy's floating hospital ship, will not be deployed until Sunday of this week.
  • The USS Bataan, a 844-foot ship designed to dispatch Marines in amphibious assaults, has helicopters, doctors, hospital beds, food and water. It also can make its own water, up to 100,000 gallons a day. And it just happened to be in the Gulf of Mexico when Katrina came roaring ashore… . But now the Bataan's hospital facilities, including six operating rooms and beds for 600 patients, are empty. A good share of its 1,200 sailors could also go ashore to help with the relief effort, but they haven't been asked. The Bataan has been in the stricken region the longest of any military unit, but federal authorities have yet to fully utilize the ship.
  • Local government is telling people NOT to call into the city because if the cellular service gets overwhelmed it could hamper rescue and recovery operations.
  • From what I’ve been able to read and hear at press conferences, the levee fix was delayed because finding the actual levee break and determining the severity took time. Then they’ve got to organize helicopters dropping 3000 lb. sandbags.
  • Ya’ll…… the governor has just announced that NO must be COMPLETLEY evac'd. There is no fixing the levee right now and they are expecting Orleans parish to have 12–15 feet of water by tomorrow.They are evacuating EVERBODY… .EVERYBODY… EVERYBODY… .They are working 24/7 to resuce people from attics and roof tops.
  • Hopefully it will make you feel better to know there are 125,000 national guardsmen coming, as well as 3 Navy ships (One is the Comfort, a giant floating hospital) and several Air Force planes.

Question about looting Bloggers wanted more information regarding reports of looting:

  • Since they’re having looting problems at the Hospitals, do they have a protocol in place to secure the nuclear medicine stuff? I'm not worried about a bunch of druggies wandering off with oxycontin, but I am worried about someone getting the materials to build a dirty bomb. I know, call me paranoid….
  • Does anyone know about the condition of the buildings on the 1000 block of Esplanade?,specifically The Italian Hall(1020). Is there any damage or has looting occurred there or in that block? Rampart borders
  • I heard about looters around the A+P on Magazine and the Rite Aid on St. Chaz. Are they looting for food at grocery stores, or are they looting houses for electronics, jewelry etc.?


Information remains a key use of media, both traditional and emerging, and was found to be key during this crisis as it was the primary means of obtaining information about Hurricane Katrina for the majority of the 2 weeks we studied.

Postofficial news Filtering was a primary function of many of the blogs in the sample, which were used to pass on official news from other sources. Some of the news originated in traditional media sources and was disseminated further by bloggers; other information was cited from press releases:

  • Water rising at 17th St. canal… Tuesday, 6:30 p.m… . Mayor Ray Nagin has announced that the attempt to plug a breach in the 17th Street canal at the Hammond Highway bridge has failed and the rising water is about to overwhelm the pumps on that canal. The result is that water will begin rising rapidly again, and could reach as high as 3 feet above sea level. In New Orleans and Jefferson Parish, that means floodwaters could rise as high as 15 feet in the next few hours. Nagin urged residents to try to find higher ground as soon as possible.
  • Nashville is now under its first ever tropical storm advisory. The remanants of Katrina are due here tomorrow.
  • NEW ORLEANS (Reuters)–Authorities suspended an evacuation of New Orleans on Thursday after a reported shooting at a U.S. military helicopter……
  • Noah's Wish, a not-for-profit organization that work exclusively to rescue and shelter animals in disasters is rescuing animals from evacuated homes in Slidell, Louisiana… . PET OWNERS WHO LIVE IN SLIDELL NEED TO COME TO HERITAGE PARK TO REPORT A LOST OR STRANDED PET. Noah's Wish staff will be at Heritage Park from 8am to 7pm everyday for at least the next 3 weeks.
  • Mike Brown says he has resigned as director of the Federal Emergency Management Agency, CNN reports.
  • The Mexican Army made history today when its convoy with hurricane relief entered Texas—bound for San Antonio. It's the first Mexican military unit to operate on U.S. soil since 1846 and the first Mexican disaster aid mission to the United States. About 45 vehicles are part of the entourage bound for San Antonio, where about 5,000 Hurricane Katrina refugees are being housed.

Providing information about damage Information about damage ranged from personal notes about particular neighborhoods to filtered information from traditional media sources:

  • I live in Mid City and my elevation is about 1.5 feet below sea level, meaning 4–5 feet of water over there, most likely. Bad, yes. But maybe the possessions that I put up on high closet shelves before leaving will still be dry.
  • Just heard from a stalwart who rode it out in his Bywater house, on Dauphine St. between Clouet & Louisa. There is NO water on Dauphine right now.
  • The scope of the disaster keeps unfolding. They say between 10,000 and 20,000 people have been pulled from roofs now (up from initial estimates of 1200 people stranded on roofs).
  • Two space shuttle facilities have been damaged by Hurricane Katrina and hundreds of workers are homeless.
  • More than 389,000 homes and businesses in the state remained without power Friday—339,237 Entergy customers, and about 50,000 Cleco customers, according to the companies' Web sites.

Checking on status of loved one's area Some posts simply wanted information on how particular locations had fared:

  • If anyone knows… what happened at the Bethany Home after the storm, call me at any time. My prayers are with you all. I have a friend in New Orleans prison. Does anyone know what happened to the prisoners? Or how the actual prison held up during Katrina?
  • Does anybody knows if Terrytown is flooded?

Providing information about missing that had been found The majority of these posts are follow-ups to earlier posts, especially “looking for missing.” Examples of this include—

  • Name is OK, I have spoken with her and she evacuated to MS, her mother and sister also evacuated.
  • This is an update to a request for information on my family they have been resued and are in AUSTIN,TX at the convention center. There names are Name1, Name2, and Name3 of New Orleans. Also to my cousin in Opelousas Name, everyone's okay!
  • Just wanted to let those of you who were also worried about him… I heard from Name last night! He apparently floated from Mid-City down to lower Decatur, Huck Finn style, and is safe at his favorite watering hole with Name and firearms.

Some of these posts were accompanied with firsthand accounts from the person who had been found:

  • Hi everyone,
  • I’ve been too tired to write today, but I wanted to pass on this email I got from Name1, with whom I work. (He does all the work on finances and book sales for Organization and Sister Name's work.)
  • We’ve been waiting and waiting to hear from him, knowing he was stuck in the French Quarter. It was with huge relief that we got this email this afternoon.
  • Name2

After harrowisng/horerific days at Omni [the Omni Royal Orleans hotel in the French Quarter] etc. many war stories to tell, but I’ll make it short at this time but will bore you with them later. On Tues Morn. 6am my aunt got up to go to bathroom, with flashlight, and fell. We did not discover her until 8, on floor bloody arm and possible hip or ribs broken. No emt service so we set out for charity hosp. were moved to superdome, they wanted to evacuate to BR [Baton Rouge] without me and we refused–sat there for 7 hours without any one seeing her, no food or bathroom opportuites–I walked around until I found a used bed pan and rinsed it out with bottled water. Finallly they said they were going to evacuate her to a new location at Interstate and Causway Blvd. [this is an outdoor location where thousands of people are congregating with medical triage] I told her this might be the best since the alterrnative was to stay at the dome with no medical help and no food etc. She agreed and they put her in the back of a truck and of course that is the last we have heard from her. I'm sure she has been moved to who knows where.

After she was taken I walked thur water,, sometimes pocket high, back to the Omni. I did hitchhike part of the way on side of Army vehicle. Got to Omni and was told by police they had taken over the hotel and I could not enter. I mentioned my family, wife, daughter, son in law, granddaughter and they said every person in hotel was evacuated, led out of the city by police escort. I told them I had just walked back from superdome, I was exhaused and I was sitting on the steps until I could decide what to do. For a brief moment I thought the best solution was to get arrested. Anyway, while I was sitting thinking, planning (all bad) I heard the lovely accented voice of Darryl from Scotland, second in command of the hotel–he helped get my aunt to the hospital that morning, I jumped up, he saw me we hugged and he told them I was with the hotel and we convinced them I should stay–AB justed called to come look at some TV coverage–I hadn't seen any tv all week and I'm embarrrased how minor my problems seem when I see people dying in the street by Convention Ctr.

AB let me a note with all our stuff in the room and I was able to contact her in Houston with a friend of my daughters. Yesterday around 1pm police ordered me to leave and escorted me out of town.

Asking about damage, no tie to area People with apparent knowledge of the area, but not necessarily any personal ties, queried about the impact of the storm on specific locations:

  • Does anyone know what the water level and damage is around jefferson and st. charles or around the 5000 area of Dryades street in uptown? Email any info
  • Does anyone know what areas in Kenner are flooded?

Checking status of one's home As one would expect, the posts for this topic, although differing in wording, were quite similar in content and the focus on how one's home or area faired during the hurricane:

  • The shotgun houses were mostly built before 1905, on high ground. The TV shows flooded bungalows (built between 1904 and 1940) and ranch houses built after WWII… but I have seen NO flooded shotguns. I live in a half double on Hagan, near the end of Bayou St. John… and worry about neighbors who stayed… plz call Name at xxx-xxx-xxxx if you have info or even a guess. THANKS IN ADVANCE…
  • live off Lake Ave in Metairie on the Jeff Parish side of the 17 st canal. Anybody got a clue about water, electricity in my area. I live about 6 blocks off Vetern's hwy toward the lake. Let me know if you know anything. All I cann see is aerial photos of my area and it looks pretty dry. Thanks Name

Providing information to help locate missing People were frequently referred to resources that could help in connecting them with missing relatives:

  • I wanted to point out the following site that has been set up today. It organizes people by primary phone numbers. You can leave a message that you are ok or look up people by phone number. Hopefully it will help. Best wishes from Houston.
  • Also check out to register a missing loved one to be rescued. The phone numbers are busy but the web form works.
  • Additionally, if you need to report a missing person, Atlanta's Red Cross office is handling requests at 404-870-4440.
  • Also, I havent been able to contact any other 504 area code number by cell or land line but I have been able to communicate with other friends via text message on the cell phone. If you have that option on your phone consider that.

Providing information about how to get help A number of posts fell into this category, fulfilling a traditional “linking” role of blogs and providing helpful tips on where to find assistance. Typical posts were as follows:

  • I have heard there are shelters in Tupelo, Miss. that are almost vacant. If anyone is seeking shelter in that area, I would urge you to contact the Red Cross there for help.
  • The Humane Society of Northwest Louisiana has partnered with LSU Shreveport to create an emergency shelter for animals from the evacuation areas. For more information about them, go to
  • For people looking for places to go check


While the helping function is related to the other areas of the Venn diagram, and therefore tied to traditional uses and gratifications, it really represents something conceptually unique in many ways. Online communities introduced the idea of “fostering community” but the “helping” function goes beyond making a connection to others and shows the humanity in an inhumane situation.

Providing social support and attempts to foster community These posts ranged from pep talks directed at the citizens of the Gulf Coast to admonitions against placing blame when there was other work to be done. They indicate how blogs can support and bring people together who may be separated by physical distance:

  • This is about rebuilding a community! What pieces and patches as we can locate, wherever they are, we need to connect them, virtually and eventually physically, until there is enough critical mass for whatever size community is left, to heal and flourish within its (newly limited) potential.
  • We are with you in spirit and will be here for you.
  • Your fellow americans are with you all the way.
  • I will continue to do everything I can for all of you in the gulf coast! YOU WILL SURVIVE AND BE STRONGER AND MORE BEAUTIFUL FOR IT!!!!
  • All of us in Chicago are praying for you all. This is absolutely horrible and gut wrenching to watch. I wish you all a speedy recovery…
  • When you’re back on your feet Big Easy, I’ll be back for another Jazz Fest! In the meantime, I'm sending a check too. Bill, San Jose
  • All this energy wasted on blaming could be otherwise well-spent trying to help. Start a church or school fundraiser. Volunteer to go down there with the Red Cross. Donate to the Red Cross (who has been in New Orleans from Day One). Just don't blame anyone else unless you are there up to your knees in filth without food or water. … so I hope that all of us will bolster each other up and remember that many of us could be facing what We can accomplish so much more by cooperating with each other to figure out how fix this. Not why. How.
  • Why waste energy placing blame? Wouldn't our thoughts be better spent on coming up with solutions to solve the problem?the victims of Katrina are facing now.
  • … for now we need to pull together. The test of a GREAT PEOPLE AND A GREAT NATION IS HOW WE RESPOND.
  • You should all get off the computer and off your butts and work to help!!! Talk is cheap!!

Offering other forms of assistance Many bloggers expressed a desire to provide personal help to hurricane victims, often in very direct ways, including offers of housing, in-kind donations, and manual labor. Still others offered to contact the loved ones of hurricane survivors and evacuees via phone or e-mail:

  • If anyone needs phone calls made to family or friends, please email me and I will call for you. I am sure you have someone worried about you and if I can help in any way, please let me know. You can email me at I know that it is not much, but I want to do what I can to help! God bless you all!
  • My family would like to open our home to a family who has been displaced by hurricane Katrina.We are located in Midtown Atlanta, GA. We can accommodate a family of five. My teen son is allowing his room and full size bed to be used and we can look into air beds and/or cots for children. All five can sleep in his room relatively comfortably. We also have two full baths. We can help with food and toiletries. for further information please contact me at
  • I want you to let you know that THERE IS A PLACE where they can come for a safe place to live temporarily. My husband has access to over 300 acres available in southwest Virginia for refugees of Katrina to camp. We are poor people and cannot afford to feed all that may come. I can try to help find temporary employment to help folks get back on their feet.
  • I am an instructor at Panola College in Carthage, TX. We are located approximately 50 miles from Shreveport. We are a two-year college offering the first two years of under graduate study in the liberal and fine arts. It is my understanding that we have 100 beds available for college students displaced by Katrina. Please contact our admissions office for more information….
  • Maple Care Long Term Care Facility in Maple Heights, Ohio would like to offer assistance to those affected by Katrina. We have space available for those seeking housing, we have mattresses, walkers and other items we would like to share. Please call Name at xxx-xxx-xxxx or email at with interest or questions….
  • I have about 600 brand new, big, pretty darn nice diaper bags that I'd like to donate to people who need them but I don't know how to go about it. If you can point me in the right direction, please e-mail me at Thanks and God bless.
  • If any women/young women victims of the hurricane need clothes, I have some I can send to you. Mostly size small shirts, size 4 to 6 in pants/shorts/dresses, and size 7 shoes. They are in very good condition. If you or someone you know could use these clothes, please send me an email to me at with your address or location and “clothes” in the subject line and I will do my best to send them right away. The relief organizations aren't accepting clothing donations, but I can see on TV that people need them.
  • I am 57. retired Fed. employee, have had birds all my life. I am more than willing to meet the needs/adoption of a bird/birds. I have had parakeets, cocktiels, African Gray, and now own a cockatoo. Please let me know if any of these avian friends need a home… . I am willing to be a permanent mother, or be a temporary mother in the case the owner wants to retrieve their avian friend. I am willing to pay/donate $$$ for the transportation of the avian friend that needs a home.
  • I'm quitting the day job, putting the small amount of personal belongings that I have into storage, and heading off to the New Orleans area to help in the aid and relief efforts that are beginning to take place…

Organization of assistance These blog posts were very diverse but the aim was to let people know how they could help (e.g., donating items or money, spreading the word, physically helping, etc.) or to make suggestions for relief efforts. There were posts that listed items that could be donated to local shelters or churches who had taken in displaced people.

  • I live in L.A. and I would love to organize a concert out here where people trade diapers, baby formula, jugs of water, and clothes for a ticket, and we’ll listen to the music of the South and pray. I was at the Long Beach Jazz and Blues festival last Sunday and there was a feeling of bitterness in the air that the cradel of the blues was hurt so deeply. There was almost a sense of desperation to appreciate blues and jazz out of defiance of the horror. I want a benefit concert here in L.A. Anyone with advice or guidance please contact me. God bless.
  • Memphis government and visitor's and convention bureau are holding a meeting this morning (Wed.) to coordinate long-term relief for the refuggees here—schools, housing, food, etc.
  • I just heard on the radio that the Astrodome was looking for volunteers to man the food lines.
  • The animal rescue organisation Noah's Wish is mobilising to send 100 volunteers to the gulf coast and rescue as many pets that have been separated from their families.
  • If anyone can take an animal or several or know anyone who can, contact the Louisiana Vet School.
  • Call your local Red Cross, or animal shelter and donate $1.00. If 1,000,000 people would donate just $1.00, that would be $1,000,000. Now just think about how many people are in the world. Think about how many people are just in New York city alone. I see people spend 3 or 4 dollars on a pack of cigarettes. So donate the price of one pack to help those victims.
  • Help spread the word about a new blog Please post on your blog and encourage your readers to help.
  • LSU Katrina Volunteer Hotline at xxx-xxx-xxxx. Louisiana Dept. of Health and Hospitals has listed (xxx)xxx-xxxx as a contact number for prof. medical volunteers.
  • I’ve formed a team for the walk, called Bloggers for Animals, and I'm hoping that I can pull together a group of Boston-area bloggers, blog-using educators, technology activists and concerned citizens to register for the walk and help raise money for the event.
  • Times-Picayune managing editor Peter Kovacs says that many print and broadcast journalists have made Baton Rouge a base of operation and what they need more than anything right now is a place to stay within about 30 miles of the city. If you can provide journalists with a place to stay–or know someone who can–e-mail your contact information to and we will pass along the message.
  • If you are interested in organizing a local effort to help schools in need, please review the resources and requests posted here. We will be updating this site with more school-related needs in the coming days and weeks.

Organization of rescue help The majority of these posts took the form of providing information about people who needed to be rescued so that those who had the ability could find survivors:

  • Elderly great aunt and great uncle in upper 80's. He needs oxygen for emphysema. Address is Street Address. in the 9th Ward. Did not evacuate to son's house, and no one has heard from them. Please send rescuers to check on their status in case they remained in their home. They have plenty of food in the house so could be still alive if stranded there. THANK YOU!
  • My uncle, Name, known as Nickname, is a WWII vet, who has cancer and is on chemo. His two-story house is located between Street and Street behind Nashville (in the area of S. Claiborne and Fontainbleau). He needs to be rescued and brought to the hospital immediately as he is probably really dehydrated. His 80-year-old neighbor is with him. Please help or notify me if you know of anything:
  • Dogs stranded at Street Address in Mid-City: Two dogs and Name are stranded at this address and desperately need any help which may be provided. Coast Guard will currently not rescue pets, and Name will not leave without dogs. If anyone may be able to provide assistance, please call (xxx) xxx-xxxx, (xxx) xxx-xxx or (xxx) xxx-xxxx.

Offer rescue help

  • Liferafts available, Name, xxx-xxx-xxxx, e-mail:… I have a company in Harvey La. we have 5–6 inflatable liferafts in stock which can be used to help evacuate. these rafts hold approx. 15–25 people each. pleas contact the above. this note is for officials only
  • Anyone needs a driver or drivers to bring relief materials water food equipmentor anything else please call xxx-xxx-xxxx, xxx-xxx-xxxx, or xxx-xxx-xxxx


The level of emotion was often evident in the postings included in earlier sections of this paper. Given that this has been a consistent although minor theme in previous research (Boyle et al., 2004; Procopio & Procopio, 2007; Riffe & Stovall, 1989), we wanted to highlight some additional details. About 35% of the bloggers expressed a discernable emotion. The most common emotions expressed (in order from most to least common) were interested/concerned, disgusted, angry, afraid, and hopeful. Examples included “… i am concerned due to some of the violence that i have heard of happening. but they feel like it is something they need to do… ,”“as a member of the military, I am disgusted at what I am seeing on here, and what I am hearing on TV… ,”“… I am so angry about the looting!..,”“… I sensed that there was an open door to get out of town and that it might not stay open long. I was a little afraid of the kind of situation I could find myself in a week later if I didn't be decisive…” and “I read NOLAs wednesday's blog about flooding on Esplanade Ave and the inability to evacuate the Bethany House before the storm but I am keeping my hopeshigh and hopefully for my mother, aunt, cousins, and all the family as well.”


This analysis looked at the functions of blogs that not only predated Hurricane Katrina but also those that were created specifically for this particular crisis. Existing blogs, such as the New Orleans Times-Picayune's (, were subsumed by the crisis almost exclusively during this time, resulting in very similar content on both established and new blogs during the period studied. The timing of the posts supported that the blogs were being used for pragmatic individual functions as opposed to commentary or other nonemergency functions because the greatest number of posts were in the first 7 days after Katrina made landfall in New Orleans. This flurry of activity also corresponded to many of the subsequent events following Hurricane Katrina—levees breaking, looting, and evacuations.

Examining the characteristics of the bloggers (RQ1) revealed some interesting insights. Although a large proportion of the posts did not enable us to identify their location, a small proportion of those identified were from New Orleans. Increasing perpetuation of technology such as BlackBerry's may change this in the future and allow people who are in need of rescue the means for communicating more directly with potential rescuers. Although the blogging examined in this study was overwhelmingly U.S. driven, there was considerable international support and concern as well.

As investigated by RQ2, the major functions of these Hurricane Katrina blogs corresponded with those found in previous blog literature including both filtering and linking about rescue needs and efforts, missing persons, ways to offer and find assistance, fostering community, and providing information on damage, government response and related stories. Additionally, the thinker function was fulfilled in numerous categories where bloggers expressed opinions, especially in comments on government response. It was interesting that although our process was very inductive several of the four main functions also correspond to previous uses of both traditional and nontraditional blogs, such as information and uncertainty reduction. This indicates that the uses and gratifications of media still remain somewhat consistent despite the passage of time and changes in media format.

While the application of uses and gratifications clearly applies to crisis communication blog content and served as a useful theoretical framework, given the prevalence of information seeking, and establishing online connections/communities, an extended typology of uses and gratifications would be necessary for this specialized form of blogging to be more inclusive of all the functions found in our study and not typically included in uses and gratifications typologies (e.g., helping).

There appeared to be an additional function of blogs in general not previously delineated, which might be termed emotive or therapeutic. This function, in which the blogs almost fulfilled the role of psychologist or grief counselor, is distinct from the opinion-oriented thinker function. Previously blogs have been found to have an information or social function, thinker or linker, whereas the emotive or therapeutic function found here provided a means for individuals to discuss their emotions associated with shared evens, in this case a hurricane. The emotive or therapeutic functions were captured throughout categories, but particularly in providing social support and sharing life experience relevant to the hurricane. This new function is most likely to be found in blogs involving shared crises (e.g., natural disasters, bombings, shootings, etc.) but may also be identified in blogs about politics or other topics that individuals tend to be passionate about.

Unlike the Thelwall and Stuart (2007) study that found blogs, among other “new technologies[,] were fulfilling a general rather than a personal information need: helping people to find out about the crisis rather than checking the safety of individual people (n.p.),” this research showed that blogs were used for pragmatic individual functions in this crisis, primarily to look for missing persons and to request and offer tangible support. Several key factors differentiate our study from theirs and may explain the incongruent results. First, their study set out to examine emerging technologies mentioned in blogs, rather than functions of blogs themselves. Second, examination of posts was conducted through electronic scanning, tracking key words according to frequency, while ours took a line-by-line approach that was able to capture the nuances of each message more fully.

Additionally, two of the three crises Thelwall and Stuart researched, the London bombing and the Asian tsunami, took place outside the U.S., yet U.S.-based blogs dominated the material they examined. It is doubtful that most U.S. bloggers had personal connections to victims of the overseas crises or would be in a position to offer personal assistance, such as opening their homes to those in need. On the other hand, more than half of the blogs we surveyed functioned as some type of “personal communication,” with locating the missing as well as organizing and offering rescue help and other forms of assistance figuring prominently. The posts were remarkable in the amount of personal information, including e-mail addresses, street addresses, and phone numbers, that people were willing to share as part of these communications. This may indicate several things—a trust in blogging as online communication, desperation in finding loved ones, and/or a willingness to go to any means necessary.

The blog postings analyzed in this study further explore the idea presented in previous research (Procopio & Procopio, 2007) that the Internet, specifically blogs, can help maintain and even build a stronger sense of community. During Hurricane Katrina, the New Orleans community was literally and physically ripped apart. Blogs and online communications helped provide a temporary but important means to keeping this community strong and helping give it the ability to move on and recover until more tangible results could be obtained.

To varying degrees in noncrisis times, we live in a fragmented and isolated society. Many of us are physically and sometimes socially isolated throughout our day—sitting at computers, in our car, even texting instead of calling our friends and family, etc. This is far removed from the days of borrowing cups of sugar from your neighbors. Although this study examined blogs used in times of crisis, it would be interesting for future research to examine this community-building function during noncrisis times. Maybe blogs and other electronic communication tools are the means for many to form communities that are not limited by distance or time. We hope this is not a substitute but rather a supplement to other face-to-face communities.

Hurricane Katrina is the first major crisis where blogs were widely used as a communication channel. While the myriad functions of blogs will continue to evolve, their documented impact on rescue efforts (Glaser, 2005) will certainly be revisited in future crises, and therefore blogs demand the attention of emergency risk communicators. As both a redundant technology and a vehicle to give voice and access to average people, blogs are a must-have addition to crisis communication toolkits. Although several of the functions we identified in these blogs are also represented in mass media, such as information and political communication, they serve an important function during crises when traditional mass media may not be able to provide the information available to bloggers because of limited accessibility to the affected areas. It would be interesting for future research to examine if these functions differ in other crisis situations either past (if electronically captured) or future.

Crisis and risk communicators should consider several best practices when including blogs in their crisis communication plans. First, they should be aware of existing blogs relevant to their communities or stakeholders, since these blogs may be dominated by a crisis once it breaks. Second, if appropriate, risk communicators should establish blogs in advance, including not only setting up the technology, but providing regular content to attract and retain an audience of readers and posters. Third, practitioners should take steps when establishing a blog to use keywords and provide meta tags that will allow search engines to find them easily; often these tags are part of the set-up template on some of the most popular blog infrastructure sites. Fourth, blogs should be part of any communicator's distribution list for press releases and other announcements, taking their place next to journalists, partner organizations, and stakeholders in receiving communication messages. Fifth, blogs should be regularly monitored during a crisis, not only in the traditional mode of public relations surveillance, but also as an important alternative communication channel that can aide rescue and recovery efforts.

Over one-third of the posts included clear expressions of emotions (RQ3). This supports an emotive/therapeutic function which will be discussed further. A great deal of people felt the need to publicly express how they were feeling about these tragic events. The most commonly expressed emotions (interested/concerned, disgusted, angry, afraid, and hopeful) ranged greatly and indicated the mixed feelings surrounding the events.

As with any study, while we paid great attention to the details there are inevitably limitations that must be considered when examining and applying the results. In hindsight, the definition of blogs has changed since the study began in 2005 and will likely continue to evolve. Application of these results should keep in mind our broad definition. Another limitation of the current research included the sheer volume of blogs and blog posts during Hurricane Katrina and the challenge of finding a truly representative sample. However, our straightforward description of the sample allows the reader to decide. Additionally, the amorphous quality and variety of blog posts makes it difficult to fit them neatly into mutually exclusive categories for coding. This study utilized content analysis as its primary method but the analysis was qualitative and inductive in nature as described throughout the paper.

Finally, the nature of content analysis does not allow for knowledge to be gained beyond the text available. We can not determine the end result of the blog posts that were analyzed. Did the therapeutic function really help people cope? Was the information gained successful in rescuing people, launching fundraising, etc.?

Future research could directly address this by attempting to contact people involved in the blogging during Hurricane Katrina and determining what effect the blog efforts had. In the event of future crises, it would be helpful to monitor how the functions developed in this paper differ by type of crises or simply evolve over time. Technology is always changing and it will likely impact the functions that blogs or other Internet communications have in times of crises.

In addition, future research may want to employ the even broader definition of social media for future studies. Not only have blogs changed since we conducted this study but social media has expanded rapidly and will continue to as technology and communication find new ways to bring people together through Tweets, Wikis, and Flicks.

About the Authors

Wendy Macias is Associate Professor in the Department of Advertising and Public Relations, Grady College of Journalism and Mass Communication at the University of Georgia. Her research interests include online health communication issues, direct-to-consumer pharmaceutical advertising and consumer processing of health communications and advertising.

Address: Department of Advertising & Public Relations, Grady College of Journalism and Mass Communication, University of Georgia, Athens, GA 30602. Email:

Karen Hilyard is an Assistant Professor of Public Relations in the College of Communication and Information at the University of Tennessee. She researches risk, health and crisis communication, particularly where the three areas overlap related to disasters and diseases.

Address: 476 Communications Bldg., The University of Tennessee, Knoxville, TN 37996. Email:

Vicki S. Freimuth is a Professor in the Department of Speech Communication and Grady College of Journalism and Mass Communication as well at the Director of the Center for Health & Risk Communication at the University of Georgia. Her research focuses on the role of communication in health promotion both in the U.S. and in developing countries.

Address: 701 Coverdell Center, University of Georgia, Athens, GA 30602. Email: