The 2004 hurricane season was the worst in Florida’s history, with four hurricanes causing at least 47 deaths and some $45 billion in damages. In order to collect information on the demographic impact of those hurricanes, we surveyed households throughout the state and in the local areas sustaining the greatest damage. We estimate that one-quarter of Florida’s population evacuated prior to at least one hurricane; in some areas, well over half the residents evacuated at least once and many evacuated several times. Most evacuees stayed with family or friends and were away from home for only a few days. With continued population growth in coastal areas and the apparent increase in hurricane activity caused by global warming, the threats posed by hurricanes are rising in the United States and throughout the world. We believe the present study will help government officials plan more effectively for future hurricanes evacuations.

The 2004 hurricane season was the worst in Florida’s history, with four hurricanes causing at least 47 deaths and some $45 billion in damages. In order to collect information on the demographic impact of those hurricanes, we surveyed households throughout the state and in the local areas sustaining the greatest damage. Using these data, we estimated that one-quarter of Florida’s population evacuated prior to at least one of the hurricanes; in some areas, well over half the residents evacuated at least once and many evacuated several times. Most evacuees stayed with family or friends and were away from home for only a few days. In this study, we summarize the results regarding the number of evacuees, types of lodging, and number of days spent away from home for the state and the regions hit hardest by the hurricanes. Using logistic regression analysis, we analyze the factors affecting evacuation decisions. With continued population growth in coastal areas and the apparent increase in hurricane intensity (and perhaps frequency) caused by global warming, the threat posed by hurricanes is increasing as well. We believe the results of the present study will help federal, state, and local officials deal more effectively with this threat.

Abstract:
This article estimates the variation in personal network size, using respondent data containing two systematic sources of error. The data are the proportion of respondents who, on average, claim to know zero, one, and two people in various subpopulations, such as "people who are widows under the age of 65" or "people who are diabetics." The two kinds of error—transmission error (respondents are unaware that someone in their network is in a subpopulation) and barrier error (something causes a respondent to know more or less than would be expected, in a subpopulation)—are hard to quantify. The authors show how to estimate the shape of the probability density function (pdf) of the number of people known to a random individual by assuming that respondents give what they assume to be accurate responses based on incorrect knowledge. It is then possible to estimate the relative effective sizes of subpopulations and produce an internally consistent theory. These effective sizes permit an evaluation of the shape of the pdf, which, remarkably, agrees with earlier estimates.

This article presents a model using data from 205 telephone surveys conducted in the same survey lab over a three-year period. The model demonstrates that while response rates are partly a function of variables reflecting effort, they are also affected by contextual variables often not under the survey vendor’s control. Significant factors that affected response rates included the salience of the survey to the population, the survey length, the type of sample (listed vs random-digit dialing), minutes per piece of sample (effort), and the amount of time the survey was in the field. A ten-minute increase in survey length results in a 7%decrease in the response rate. An increase of one day in the field per one hundred cases (fielding time) results in a 7% increase in the response rate. An increase of one interviewer minute devoted to each piece of sample released results in a 2.2% increase in overall response rates and a 3.4% increase in random-digit dialing response rates.

In this study, we describe an approach that can be used to estimate the demographic impact of hurricanes and other natural disasters, provide a detailed assessment of the 2004 hurricane season in Florida, compare the 2004 hurricanes with Hurricanes Andrew (1992) and Katrina (2005), and draw several conclusions regarding the likely impact of hurricanes and other natural disasters on future population growth. This study provides the most comprehensive assessment yet of the 2004 hurricane season in Florida and adds to the small but growing literature on the demographic effects of natural disasters.

By most measures, the 2004 hurricane season was the worst in Florida’s history. Four hurricanes blasted through the state between August 13 and September 25, with Charley making landfall on the southwest coast near Punta Gorda, Frances on the southeast coast near Stuart, Ivan in the panhandle near Pensacola, and Jeanne nearly retracing the route followed by Frances. This was the first time in recorded history that four hurricanes had struck Florida in a single year. Most parts of the state were hit by at least one of the hurricanes and some were hit by two or even three. Overall, the storms were blamed for at least 80 deaths in Florida and caused more than $20 billion in damages.

In this issue of Florida Focus, we discuss some of the major findings of surveys conducted by BEBR of permanent residents living in Florida when the first of the hurricanes struck.

Florida was struck by four devastating hurricanes in 2004. In a previous issue of Florida Focus, we presented the results of a survey designed to assess the impact of those hurricanes at the state level. Those results provided a useful statewide overview, but provided no information on the impact of the hurricanes on local areas in Florida. Since some of the most populous parts of the state were largely unaffected by the hurricanes, state-level results understate the impact of the hurricanes on many local areas. In this report, we describe the results of surveys conducted in the areas most strongly affected by the hurricanes. These surveys, funded by the Florida Legislature, provide a wealth of information on evacuations, housing damage, population displacement, and reconstruction. The results of these surveys deepen our understanding of the full impact of Florida’s 2004 hurricane season.

Several algorithms and software packages have been developed for displaying the relationship between actors within a whole (sociocentric) network. These visualization packages use as input an adjacency matrix representing the relationship between actors, and have occasionally been applied to personal (egocentric) network data. Personal network adjacency matrices require respondents to report on all alter-alter ties. This is an enormous respondent burden when the number of alters goes much beyond 30. We report here on an effort to reduce that burden by having respondents build their own personal networks, interactively, on the Internet. In a study on smoking, 100 respondents (50 smokers and 50 non-smokers) listed 45 network alters and provided data on whether each of the 990 pairs of alters talked to each other. We used a program called EgoNet to collect these data. Fifty of the respondents (25 smokers and 25 nonsmokers) then completed a similar exercise over the Internet, using a visual interface, called EgoWeb. There are clear mode effects on personal network composition and structure.

Although response rates have always been important in survey research, they have become central in discussions about survey quality during recent years. It is not unusual for clients, particularly those from state and federal agencies, to have expectations about the response rate that they will get on a survey. At the same time, the number and types of surveys have ballooned, yielding a combination of modes, populations, and topics that have increased the complexity of both data collection and analysis. This has caused some confusion as clients, who have little concrete data for comparison and sometimes little experience in survey research, develop expectations for response rates that are out of line with current trends for the type of survey they commission.

During the next decade, many African countries will almost certainly become more politically and economically stable. An increase in the amount of foreign private investment as well as the blossoming of local entrepreneurship will inevitably follow. When this happens, government and private institutions will face obstacles of a different kind. As in most developed economies, the focus will change from building roads to building an information infrastructure. Success and failure of businesses in developed economies is largely based on who has access to the most current and accurate information. This information may range from price-level data to market demand surveys.

Social network analysis (SNA) is both a theoretical perspective and a set of methods. In terms of theory, SNA extends and complements traditional social science by focusing on the causes and consequences of relations between people and among sets of people rather than on the features of individuals. In terms of method, SNA focuses on the measurement of relationships between people. By quantifying the relationships between people, network analysts can apply models and techniques that are commonly used across the social and natural sciences.

Most personal (egocentric) network studies describe networks using measures that are not structural, opting instead for attribute-based analyses that summarize the relationships of the respondent to network members. Those researchers that have used structural measures have done so on networks of less than 10 members who represent the network core. Although much has been learned by focusing on attribute-based analyses of personal network data, the application of structural analyses that are traditionally used on whole (sociocentric) network data may prove fruitful. The utility of this approach becomes apparent when the sample of network members elicited is relatively large.

In this paper we compare two methods for estimating the size of personal networks using a national representative sample of the United States.

In this paper we introduce and evaluate a method for eliciting a representative sample of total personal networks. First names were used as a cue to elicit a sample of 14 alters from 712 respondents through a telephone interview. Network characteristics for each respondent were calculated as averages and proportions across the 14 alters. These were compared to other studies using more specialized network generators. Our method produced results which are logically consistent with those expected from a generator that elicits a sample from the total rather than a specialized subset of the total network. The proportions of kin relations, average tie strength and frequency of contacts are found to be lower than network generators designed to elicit networks of social support. Given our conclusion that the sample is representative of the total network, we examine the varying characteristics of respondents and their networks based on the domination of a particular relation type in their network. This analysis provides answers to such questions as 'What characteristics of respondents account for the proportion of family relations in their network?' and 'What are the similarities between respondents whose networks are made up of mostly work-related relations?'

Many studies have considered the economic, social, and psychological effects of hurricanes, earthquakes, floods, tornadoes, and other natural disasters, but few have considered their demographic effects. In this paper we describe and evaluate a method for measuring the effects on Hurricane Andrew on the housing stock and population distribution in Dade County, Florida

This article raises some questions about the use of the word "know." What exactly do we get when we generate ties in this way? Is knowing itself multiplex? And what is the measurable error which we can expect when knowing is used to define network ties?

In what follows, the researcher summarizes three studies that shed some light on this question. The first is an in-depth descriptive study that focused on defining "differences in knowing" of people. In the other two studies, these differences were tested with representative samples.