WORKFORCE 3ONE TRANSCRIPT OF WEBINAR Natural Disaster Impact Assessment: Using OTM for Labor Market Analysis THURSDAY, MARCH 19, 2009 Transcript by Federal News Service Washington, D.C. BRIAN KEATING: All right. Well, with that I’d like to turn things over to Anthony Dais from the Employment and Training Administration. Anthony, take it away. ANTHONY DAIS: Good afternoon, everyone. I don’t expect you to say “good afternoon” but I hope you did that back in your offices. I’d like to welcome you to today’s webinar, Natural Disaster Impact Analysis: Using OnTheMap for Labor Market Analysis. This is the first Workforce3One win-win community of practice webinar. The very first one. Thank you for joining us. In this webinar in the win-win network we’re going to launch a series of activities that show you how to do things, not just demonstrating a system. This webinar is being recorded so that those folks who weren’t able to join us today will be able to download this or view it online on Workforce3One and in the community of practice, which we’ll talk about later. Now, our presenters, George Putnam and Allan Ross, have been doing a great job working with the Local Employment Dynamics program for many years. I will tell you about their background a little bit as I introduce them. But I’d like to – and I’d like to let you know a little bit about the focus here. The focus is on an event that occurred across state lines – and they’ll get into great detail with it – and it was a natural disaster event. But I want you to think about other kind of events that this same approach could be used to help you gain a greater understanding of the impact on the work and on people. Continuity of operations, planning for pandemic influenza, that’s another great example of how you can use this tool. A natural gas leak – and that’s not a natural disaster but it can be a disastrous situation for many. Or power grid failure somewhere. Or a community experiencing large numbers of layoffs in an area or a big swath could use these same approaches to gain greater insight for planning and investment decision making. This tool is really a great tool – the Local Employment Dynamics tool. It’s available to you 24/7 – 24 hours a day, seven days a week. It’s prepaid by the government and so it’s available to you. And it’s on the Internet. So as long as the Net is up, you have access. I won’t say anymore about it because George and Allan will give you a great presentation. So let me tell you a little bit about our presenters. George W. Putnam is the assistant director of Economic Information and Analysis for the Illinois Department of Employment Security. He’s also held appointments at the Argonne National Laboratories and the University of Illinois at Chicago. His areas of specialization includes labor market forecasting, industry and occupational statistics and survey research. He has worked with the Local Employment Dynamics since basically its beginning in 1999. Allan Ross – Allan B. Ross – is a manager of the Local Employment Dynamics program for the Illinois Department of Employment Security. He has worked in the LED program since 2004. He is a trainer on the OnTheMap application at national workshops and in his state, and prior to this position Allan was an information technology analyst and developer in the telecommunications and finance industries. I’m just going to not say anything else but turn this right over to George Putnam. Thanks, George. Are you muted? GEORGE PUTNAM: – Tony. As Tony mentioned, the focus of this presentation is labor market information in disaster assessment. It’s based on the Disaster Assessment report, which is available for download in the file sharing window of this presentation. Before we get too far along I want to make sure to recognize Matthew Graham at Census and Len Preston in New Jersey and Tim Smith in Missouri for their excellent kind of reviewer comments at several stages of this project. The first question that we want to look at today is: Does LMI have a role? That is, in the array of uses of LMI does it have a niche in disaster assessment? And if it does have a role, I want to name the specific ways in which OnTheMap can be used to sort of interface LMI with disaster assessment efforts; that is, what are the critical pieces of information that are afforded by OnTheMap – or OTM – to assist disaster assessment? And does this information have a recognizable sort of value-added for the purpose of disaster assessments? In the second part of this presentation Allan will provide sort of an abbreviated step-through of how OnTheMap can be used in disaster assessment. The report that’s available for download is the basis for Allan’s sort of step-by-step process with particular focus on the features and capabilities of OTM. But because of time limitations and sort of the medium of this presentation, Allan’s going to focus on kind of the information flow of the OnTheMap application more than the sequence of key strokes, which you’ll find in the report itself. The third section I will return and discuss the potential of OnTheMap to advance our ability to estimate workload from disaster-related initial claims and to facilitate service delivery. The LMI staff in Florida have been very helpful in providing good examples of their challenges in this area that have helped shape much of what I’m going to present. The limitations of traditional labor market information severely constrain its utility for disaster assessments. In the past, labor market information has been mostly available for larger geographic configurations. Those of us who work in this area are quite aware of this limitation, it’s focus on either state level labor market information or a metropolitan area, or in some limited cases counties. Disaster assessment, though, as Tony mentioned, often doesn’t fit well with these configurations. The impact of natural disasters is not often kind of circumscribed by the boundaries of either a state, the metropolitan area or the whole county. Indeed, our example today includes parts of multiple states and parts of multiple counties. Second, labor market information is produced in accordance with state and federal mandates on confidentiality. We – that is, those of us in the LMI community – rely on employer participation in our collection of LMI – that is, the microdata – and pledge confidentiality to ensure that information on a particular establishment is not released to the public. The most common form of protecting confidentiality of LMI has been what is known as sort of the primary/secondary disclosure proofing. OTM was designed from the beginning as a statistical program to address both of these concerns. First, the unit of geography is not the state, the metropolitan area or the county, but the unit of geography is actually the census block, a constituent component of a census tract. Users can literally recreate the boundaries of the natural disaster with the OTM software that taps into a rich database of LMI for only the configuration of the natural disaster. Second, the OTM initiative implements an innovative approach to disclosure proofing using statistical imputation that permits release of unprecedented detail on workers, industries and residences while maintaining confidentiality. And I think by the end of today’s presentation you’ll appreciate more of that statement on unprecedented detail. One construct that helps me to better understand the breadth of OnTheMap information is its reliance on four basic categories of information. Within OTM you have categories of information on: worker characteristics; second, on workplace characteristics; third, on workplace location; and fourth, on worker residence location. The overarching power of OnTheMap is its flexibility to establish linkages between these categories of information. And in fact, OTM has integrated these linkages in a way that has created what I call matrices of key labor market attributes. And I’ll be talking more about these matrices as we go through the presentation. But for this part let’s just say that these matrices formulate a rich information yield. For example, the first two questions focus on age and earnings of workers in an area and reflect the information yield of linking worker characteristics to workplace location. What is the percentage of older workers or younger workers in an area? Or what is the percentage of workers earning more than $3,400 per month, versus those earning $1,200 or less? This is important because we know that the labor market behavior of older workers and younger workers differs, as does the labor market behavior high-wage and low-wage workers. The next two questions link workplace characteristics and workplace location. In the industry mix of an area and the employment concentration within that industry mix are critical to understanding the labor market infrastructure of an impacted disaster area. The final question reflects the information yield of linking workplace location and worker residence location. What is the commute pattern of workers to the disaster impacted area? We know that commutation shapes the labor market behavior of workers. And Allan’s presentation takes you through step-by-step each of these questions and responses in the form of OTM reports as it relates to the particular disaster that we’re looking at. The aforementioned matrices of key labor market attributes are themselves dynamic and not static. And by this I mean these matrices – think of them as sort of pivot tables where the rows of information of worker characteristics can become columns of groupings of workers. In the prior slide we talked about the percentage of workers in an area that are older or younger. The dynamic nature of the attribute matrices means that we can shift the focus to older workers as a distinct population of workers – that is, it’s taking that row and shifting it to a column – and generate findings on the earnings structure of only older workers, or the industry employment pattern of only older workers, or the commute patterns of only older workers. This dynamic nature of the attribute matrices permits users, such as you and me, to generation findings for groupings of workers by age, earnings and industry sector. I should like to stress that the infrastructure, tools and software in today’s presentation are made available through a unique federal/state statistical program that is housed at the Bureau of the Census, the Labor Employment Dynamics program. And through it you can access the LED site, its mapping software, and the resultant maps and data tables that are all available free of charge. Our hope is that you find the training materials at the Census site – and, in fact, the guide that’s available today – as useful presentations on a step-by-step procedure to actually use the application and the tools to enhance usability. And now Allan will move through a topical presentation of the OTM application and conclude with an analysis of the comparative information made available by this rich data source. And then last, I will return to discuss the use of OnTheMap for estimation of disaster-related initial claims and service delivery. ALLAN ROSS: Good afternoon. We are presenting a four-step approach to using OnTheMap for studying labor markets in areas damaged by disasters. And although this approach is directed toward disaster assessment, the techniques presented can be used for other types of study, as Tony had mentioned earlier. The process begins with gathering detailed real-time information about the event, principally using the Internet. (Pause.) Different types of disasters affect areas of different shapes. Hurricanes, earthquakes, and toxic gas releases affect broad areas, while tornados and floods affect relatively long, narrow, linear buffers. For events that have starting and ending points, like tornados and hurricanes, the objective is identifying the starting point, the path traveled and the end point. Background sources usually express locations in terms that the current version of OnTheMap cannot use, including place names, landmarks, latitude and longitude, and street or highway intersections. The process of how to resolve these kinds of geographic descriptions into something usable in OnTheMap is the subject of the next section. The extent of destruction or the estimated duration of disruption will affect the overall analysis of the impact on the labor market, though they aren’t required within the OnTheMap application. In a few minutes we will begin looking at a tornado area that devastated an 11-mile by one-half mile path that crossed two states and affected employment in three states. MR. KEATING: And Allan, this is Brian Keating, the webinar facilitator. If you could just speak a little louder that’d be great. MR. ROSS: There is no cookbook recipe for finding relevant background information. You have to be creative. One way to start gathering information is to directly browse one or more of the Web sites known to host that type of information. This slide lists some useful Web pages. Another way to start is simply use a search engine such as Google or Yahoo!. This presentation focuses on a study of a labor market in an area that was severely damaged by a tornado last year. The reason this event was selected for the study is that the damaged area crosses two states and affects workers who reside in three states. OnTheMap is uniquely able to work with this kind of geographic diversity. This slide has an excerpt from a National Weather Service bulletin. The National Weather Service is a prime source of weather-related information on disasters. This clip identifies the type of disaster – tornado – its date, time and duration, and some detail on the extent of the damage. The National Weather Service bulletin did not give the exact starting location, but other Web sites did name the Southaven Mississippi High School as the place where the tornado touched ground. Thus, the two ends of the tornado path are known. As – (inaudible) – the National Weather Service bulletin, the tornado lifted near the Hickory Ridge Mall in Memphis. The mall’s Web site provided additional background information on the extent of the damage and the expected length of disruption. This section focuses on using the Internet-based tools to transform the location gathered in the first section into something that can be used with OnTheMap. OnTheMap provides several methods of defining a study area. As George mentioned, areas affected by disasters do not tend to fall within defined political boundaries. That means that OnTheMap’s easy-to-use selection layer method – which enables selecting predefined geographies such as census tracts, towns, congressional districts and zip code areas – cannot be used in typical disaster studies. The area of the case study’s tornado, which was discovered in the “gather information” section, begins at a high school and ends around a shopping center. OnTheMap does not support use of place names, landmark names, addresses, intersections, or latitude and longitude. When working with such information it becomes necessary to work with a third-party tool to develop the kind of information that works with OnTheMap. The objective is to develop a study area in OnTheMap based on street or highway intersections. Our objective for resolving the location of the OnTheMap study area was to be able to draw a linear buffer starting at the high school and ending at the shopping center – an 11 mile distance with a one-half mile width. To draw such a shape, we needed to find the street names and intersections near the landmarks named in the information sources. For this study we visited MapQuest’s Web site and entered the name of the school, the city of Southaven and the state of Mississippi. MapQuest returned a map of the area with a star symbol representing the school’s location. This is part of the display that resulted. It shows the school near the intersection of Rasco Road West and Interstate 55. Then we did a second MapQuest query, entering Hickory Ridge Mall as the place name, and Memphis and Tennessee as the city and state. This is part of the return display, which shows the mall at the intersection of Winchester Road and Hickory Hill Road. So now we knew where to eventually draw the study area in OnTheMap. But before going into OnTheMap, we generated a satellite view of the area. Having that photo familiarized us with the general topography, which helped us interpret the results we finally obtained in OnTheMap. While in MapQuest we switched to the aerial map view for a combined map and satellite view, and then zoomed out to obtain this picture of the tornado path. The white line has been drawn over the map to show the route the tornado took. So even before going into OnTheMap we see that a small portion of the area is residential, a large portion is industrial, and that a fair amount of the territory appears largely undeveloped. We’re not yet ready to draw the study area in OnTheMap, but we’ll get there very soon. When we do, in Study Area Selection, this is how the screen will appear after navigating OnTheMap to display an area encompassing more than the study area and then selecting the line option with a .25 mile radius, and then using a cursor to draw a straight line between those two end points. Here you see the drawn line and the blue buffer surrounding it. Now that we have gathered background information and determined the physical location for our study, we have enough information to begin the process of using OnTheMap to generate labor market information. We just mentioned navigating OnTheMap to display the geographic area of interest. After the display has been set to encompass the study area, a number of selections are needed under OnTheMap’s analysis category. The screen lists parameters required to generate OTM maps, in the order listed. We don’t have the time to discuss each of the OTM parameters; however, the downloadable disaster assessment document that George mentioned does go into that level of detail. And there are downloadable tutorials and recorded Webinars at Census and ETA sites. I do need to mention a few of these parameters, however. A word about Years. OnTheMap currently makes data available for the second quarter of each year from 2002 to 2006; however, 2006 data for Mississippi are not available, a consequence of Hurricane Katrina which struck the Gulf Coast on August 2005. For that reason, 2005 data were used to generate the reports and maps referred to in this study. A couple of slides ago we discussed Study Area Selection. This is where you actually select the study area and where we would be drawing the 11 by a half mile buffer. Finally, the Job Type parameter needs some discussion because it’s important and must not be overlooked. Under the Job Type parameter one of four job types is selected. First let’s consider the two “all jobs” types, then the two “private sector job” types. “All jobs” literally is a count of all jobs, including both private and public sectors. When a worker has more than one job, each is counted. This job type will yield the highest job count of the four and this type is most comparable to job counts reported by other applications. For example, it is comparable to beginning-of-quarter employment. “Primary jobs” also reports jobs whose ownership is either private or public, but it only reports workers’ highest paying jobs. Additional jobs are not counted, so this count will be lower than “all jobs.” A worker can only have one job that is the highest paying, so this can be considered a count of workers. “All private jobs” omits the public sector. There are several reasons why excluding public sector employment may be desirable. Some public sector agencies may report their employment from a single location rather that at specific local offices. Furthermore, certain public sector employment is simply not included, such as federal and military workers. For that reason, I typically use private sector data in my reports whenever possible. “Primary private jobs” also omits public sector employment and it only reports workers’ highest paying jobs. So this, the second workers count, is limited to the private sector only. The case examples used in this webinar and in the downloadable Disaster Assessment document all use “primary private jobs,” so the report examples that we use are all of worker reports – that are often considered worker reports, not job reports. Having completed our background investigation and considered how to specify study criteria to OnTheMap, we will now introduce use of the reports and maps generated by OnTheMap. First let’s consider the two types of analysis. The two types of analysis listed in the left column are: area profile analysis and shed analysis. The first type of analytic report is the area profile report. This is a demographic profile of workers employed in the selection area and the demographics may pertain to either the workers employed in the selection area or the working residents of that area. The other report is the shed, which can either report where workers employed in the area live or where employed residents of the area work. First we’ll examine a couple of area profiles. We’ll start with a comparison of the demographics for workers employed in the tornado area to the area’s employed residents. We facilitated this comparison by generating an area profile of the area’s workers. And then we generated an area profile of the identical area’s employed residents. And then we pasted the two reports into Excel, positioning them side-by-side. First we see that the number of workers employed in the area is nearly five times greater than the number of employed residents – 7,100 divided by 1,500. Recall that when performing our background analysis we observed on the MapQuest aerial view that the area appeared to be more industrial than residential, and these data confirm that. This area can be said to be a net importer of workers and an employment magnet. The workers’ age characteristics show little variation between the tornado area’s workers and the area’s employed residents. Thirty-two percent of the tornado area workers are age 30 and younger, and a little less than 32 percent of the employed residents are in the same range. The middle and older ranges are also similar between the two groups of workers. The jobs by earnings paid characteristics indicate that a slightly higher proportion of the area’s workers earn less than the employed residents; 27 percent of the workers earn $1,200 per month or less, while 22 percent of the employed residents are in that range. The top three industries for both the tornado area’s workers and the area’s employed residents were manufacturing, retail trade, and accommodation and food service. Those three industries accounted for 61 percent of the workers’ jobs and 41 percent of the employed residents’ jobs. However, within those industries manufacturing provided almost 28 percent of the areas jobs, whereas only 12 percent of the employed residents work in that industry. There is a greater concentration of area workers in retail trade and a lesser concentration in accommodation and food services. Next we’ll compare area profiles of workers whose jobs are located in different parts of the tornado area. This will focus on workers employed in different parts of the tornado area, instead of comparing workers employed to employed residents. The mall area seemed to be a compact area of employment concentration and it appeared likely that the mall would have different labor market characteristics from the rest of the area. So we chose to compare side-by-side the entire tornado path, the mall and the tornado path except for the mall. We reused the report of workers employed in the tornado area; the first pair of columns entitled “entire tornado path” should therefore look familiar. Then we generated an area profile of the Hickory Ridge Mall by selecting a point at the center of the mall and opting for a study area with a one-quarter mile radius around that point. To develop the “tornado path except mall” columns we subtracted the mall statistics from the “entire tornado path” results. Indeed, we did find some interesting differences between the shopping center labor market and the tornado area outside the shopping center. First, although the mall occupies only a tiny portion of the tornado area it hosts 2,380 of 7,137 of the area’s workers – fully one-third of the area’s workers. Next we see that mall workers are distinctly younger; 52 percent are age 30 or younger, compared to 21.6 percent of their counterparts in the nearby damaged areas. Mall workers are paid less. Fifty seven point seven percent earn $1,200 or less per month, compared to 12.1 percent. So what accounts for the differences? One factor seems to be in the dominant industries located in the two areas. Eighty percent of the mall workers are employed in just two industries: retail trade, and accommodation and food services. Only 10 percent in the outlying area of the tornado path work in those industries. Businesses in these two industries tend to employ younger workers and offer lower pay. Outside the mall, nearly 80 percent of the workers are employed in manufacturing, wholesale trade, transportation/warehousing, and real estate, industries typically paying higher compensation. In contrast, less than 6 percent of the mall workers are employed in those industries. Now we’ll consider OnTheMap’s other type of analytic report, the shed report. We will view, side-by-side, the tornado area’s labor and commute sheds. The labor shed is the cities and towns, counties, and states where area workers live. The commute shed is the cities, towns, counties, and states where the area’s employed residents work. OnTheMap supports several different types of rollups. For this study we accepted the default rollups of cities and towns, counties, and states. But when entering selection criteria for shed reports, you can override the defaults and choose alternates, including congressional districts, workforce investment areas and zip codes. George will use one of these alternate rollups in his presentation on the use of OnTheMap and disaster-related claims. Most of the area’s workers – almost 60 percent – live in Memphis, Tennessee; and most of the area’s resident workers – 52 percent – are employed in this city. However, the area’s workers are more dispersed across other cities and towns, whereas 15 percent of the area’s resident workers are in a single town, Southaven, Mississippi. This map presents a network of points depicting the Labor Shed. Each point represents a range of the number of tornado area workers who live in a census tract. This map shows that area workers travel to work from a broad area. The scale at the left of the map shows the range of workers represented by points of various sizes. This map also presents a network of points, but these represent ranges of the number of tornado area resident workers who work in a census tract – the commute shed. This map shows that area’s employed residents work in a few concentrated areas. Viewed together, the maps suggest considerably differing commuting patterns for travel between home and work for area workers and area employed residents. This observation reflects the findings in the comparative shed report we just looked at, and underscores the value-added of consistent presentation of information in OnTheMap, whether the user views the tabular or graphic data. These two maps show that area workers travel to work from a broad area, whereas area employed residents work in a few concentrated areas. That completes my portion of the webinar. Now George will present an approach to the use of the commute shed report and the commute shed map for estimating initial unemployment insurance claims relating to disasters. MR. KEATING: And before we go back to you, George, I just want to point out quickly that we are going to be taking questions over the phone. So once again, you’re welcome to enter any questions or comments into that chat window. And if you haven’t had a chance to do that yet, I encourage you to do that as we go through this next segment of the webinar. However, we’re also going to be taking questions over the phone. So if you’d like to submit a comment or question over the phone what you need to do is press *1 and then you’re going to go ahead and record your name at the prompt. And that’ll put you in a queue so we can call on you when we get to that part of the webinar. All right. Very good. Back to you, George. MR. PUTNAM: Allan outlined the steps to generate demographic, earnings and industry reports for both workers of the disaster-impacted area – some of whom may reside elsewhere – and the employed residents in the disaster-impacted area – some of whom may work elsewhere. Before kind of proceeding I want to just focus our attention on two important characteristics of Allan’s presentation. First, that we’re focusing on a disaster the dimensions of which were an 11-mile long tornado path and a half mile wide. So I’d like you to make sure that you keep that image in mind that we now have a tool that can generate labor market information for a corridor that’s a half mile wide and 11 miles long across, once again, several states, several counties. Second is – I know that Allan moved quite quickly through his tables of data – and those are included in the report itself, so you’ll have some time to review those with his comments – but what I hope we were able to communicate is the flexibility with which users are able to shape the information to meet the purpose that’s intended. So Allan first showed comparisons between where workers are employed in the tornado path with where employed residents work in the tornado path. Then he looked at parts of the tornado path, the mall compared to the other components in terms of, once again, the labor market information of workers who are employed in those areas. So I want to sort of extend that discussion now to the role of labor market information in disaster assessment to the specific topic of disaster-related initial claims. One-Stop centers are a primary state actor in response to natural disasters. Disaster-related initial claims typically are the responsibility of state workforce agencies. And the effectiveness of a workforce agency response to a sudden surge in disaster-related claims depends on accurate workload estimation associated with a specific subpopulation – that is, impacted workers – and an understanding of the geography of service. Traditional LMI products often lack the granularity with respect to the subpopulation of workers and lack the specificity of relevant geography. OnTheMap bridges the gap between the needs of One-Stop centers and the deficiencies of traditional labor market information. We present an approach to utilize the unique information offered in OnTheMap: first, to estimate the workload of disaster-related initial claims; and second, to facilitate service delivery associated with disaster-related initial claims. Although obvious, I should like to stress that both of these activities – that is, the workload estimation and service delivery – should be part of a broader strategic response plan that merits on-going dialogue between the numerous state partners. My objective in this section is to provide a firmer empirical basis for decision making and policy formulation that, rightly so, should also reflect the invaluable experience of state UI officials and local office managers. Individuals who submit disaster-related initial claims are eligible for either regular unemployment insurance benefits – as specified in state and federal law – or disaster unemployment assistance – that is, DUA. Only those individuals who are not covered under regular benefits can quality for DUA, and the key requirements are that the individual must be, as a direct result of a major disaster, either unemployed, unable to reach the workplace, or unable to work due to a disaster-incurred injury. The remainder of this presentation uses the term “disaster-related initial claim” to be comprehensive of both regular UI benefits due to a disaster and DUA. The current methodology that is used by many states requested to generate an estimate of the potential number of disaster-related claims often goes something like this. They use a projected severity index that ranges from 0 percent to 100 percent, where 0 percent would be no impact and 100 percent would be full or severe impact. And this index or this percentage is usually based on historical experience coupled with analytic adjustments arising from the particulars of a specific disaster; that is, it may have occurred in proximity to a transportation artery or the flood plains. The use of Internet sources of information, as identified by Allan, is a useful supplement to this activity. The use of labor force as the core data source equates impacted population with employed and unemployed residents of the area when, in fact, only a portion of these individuals may actually work in the area affected by the disaster, which once again is a requirement for approval of a disaster-related initial claim. The flexibility of OnTheMap enables users to move easily to a redefinition of impacted population from employed residents of the disaster area to workers employed in the disaster area. Initial claims resulting from disasters cover individuals who are unable to work because their employment status has been impacted by a major disaster. The workers employed concept closely aligns with this target group. In addition, the labor force concept suffers from constraints caused by rigid geographic definitions. The federal/state statistical program that produces the monthly estimate for labor force limits its reporting to traditional political geographies such as states, metropolitan areas, counties and municipalities. These civil divisions can reduce the preciseness of geographic coverage that is so critical to disaster assessment. The redefinition of the impacted population to workers employed within the geographic boundary of a specific disaster area – as opposed to a prescribed civil division that may at best be nearby – is central to a more realistic estimation of workload size. The case study used in this report extends across the parts of several counties and, in fact, could not be estimated using the traditional labor force approach. We used the employed residents concept of OTM as a proxy for the count of employed by households that would be reported in the labor force data. The key point, which is one made by Allan as well, is that the employed residents estimate is only 1,511 persons and the workers employed estimate is 7,137 workers. In this case study the traditional approach would have severely underestimated the claims workload for the simple fact that the disaster area is a net importer of workers, an employment magnet that draws more than five times the number of workers from the surrounding area than are employed residents. A second concern for One-Stops in disaster assessment is the geography of service delivery. The flexibility of OTM makes it possible to estimate the number of workers employed in the impacted area who reside in the subject area as distinct from those who reside elsewhere, even in another county or state. A comprehensive strategy for service delivery to a person filing an initial claim requires a delivery protocol capable of serving each of these claimant pools near their place or residence. OTM reports that 7,137 private-sector workers are employed by businesses in the tornado path. If this is the worker pool then, where do they live? Referring to our earlier discussion in the introduction, this links the workplace location to the residence location. Only 48 persons reside within the tornado path and 7,089 persons outside the tornado path – well over 90 percent. This discrepancy underscores the importance of using the worker employed concept. A state workforce agency might naturally allocate most of its resources to the One-Stop center closest to the tornado zone to administer disaster-related initial claims when, in fact, most potential claimants reside outside the tornado zone. The delineation by state is not that surprising, given that the tornado path touchdown is at a Mississippi high school and it moved northeast into Tennessee, but it does demonstrate the ease with which – and necessity – OTM users can access key labor market attributes for interstate analysis. The preceding slide offers a delineation of service delivery in terms of within and outside the tornado path, or within or outside of Tennessee. While I want to emphasize that this information gets us further along a service delivery discussion than the traditional labor force approach, it does not meet the requirements for an efficient state response to disaster. OTM flexibility enables users to articulate its granular data to comprehensive state workforce formulations. Allan showed the data display of this map in one of his last slides. OTM has the flexibility to display on this map the workforce board areas delineated by lavender borders. Superimposing the workforce board configurations on the data display shows clearly that workforce board number 13 in Tennessee and workforce board number two in Mississippi have the largest number of potential claimants. OTM moves seamlessly from graphic display to tabular display, and because of its sophisticated disclosure proofing can display detailed and consistent data in both views. Nearly 80 percent of the private sector workers in the tornado path reside in workforce board number 13 in Tennessee, and another 12 percent in workforce board number two in Mississippi. We can readily identify the specific workforce boards that are likely to receive the greatest demand for its services and need to anticipate expanded service delivery in response to the tornado disaster In these last several slides we have discussed an empirical definition of impacted workers that more closely aligns with disaster-related initial claims. We’ve also demonstrated the importance of this new definition for more accurate workload estimation of disaster claims; and third, identified the allocation of this workload by workforce board. The critical final step, though, is to link this information to actual service delivery that occurs in the One-Stop centers. Can OTM move the discussion of service delivery workload to a location that has been vested with state resources to respond to worker needs? OTM not only permits the user to aggregate worker characteristics, workplace characteristics, workplace location and residence location by workforce boards, but users can also display One-Stop center locations as a layer on top of the workforce board data map display. In this example, we choose Tennessee workforce board 13 and displayed the One-Stop centers, as indicated by the yellow stars. The bulk of the centers are located in the western portion of the workforce board. The bottom southern boundary of this workforce board also represents the state border between Tennessee and Mississippi. We zoom into the western area of workforce board number 13, which also includes the residences for the bulk of the workers employed in the disaster area. This map displays the One-Stop centers in the western portion of the workforce board 13 in Tennessee, together with the plotting of residences for workers employed in the tornado-impacted area. Given the spatial display of this map, the discussion concerning efficient service delivery is probably going to focus on this area along I-55 – in the western portion of I-55 – and second in the northeast of the tip of the tornado path. OTM lends a specificity to this discussion that is unprecedented and timely and creates opportunity for efficient service delivery. Traditional LMI has not been perceived as a critical information asset in disaster assessment. It lacked granularity on worker and workplace characteristics and specificity on workplace and residence locations. OTM has elevated LMI to a central position in the arena of disaster assessment. This final map provides a spatial display of impacted workers by residence in proximity to One-Stop locations to access state resources. OTM has demonstrated its utility in the area of disaster-related initial claims. You may remember that this slide appearing in my introductory remarks. To repeat, the matrices of key labor market attributes are themselves dynamic. The rows of information on worker characteristics can become columns of groupings of workers. In the immediately preceding slides we centered on claims workload estimation and service delivery for workers impacted by the tornado. The dynamic nature of the attribute matrices means that we can also shift focus to older workers as a distinct population of workers – that is, once again, moving this category from a row to a column – and generate findings on claims workload estimation and service delivery for older workers impacted by the tornado; or, for that matter, shifting younger workers and just looking at younger workers as a group of workers; or shifting high-wage workers and looking at workload estimation of service delivery for high-wage earners or low-wage earners. The flexibility that comes with these dynamic attribute matrices is not limited to a particular use of OTM; rather, OTM was created with a vision for such flexibility. Thank you. MR. KEATING: All right. Very good. This is Brian Keating again. This is the webinar facilitator. Thank you, George and Allan. We’re just going to ahead and put your contact information up on the screen. In case anybody has any questions or comments after the fact, you have direct access to George and Allan through these e-mail addresses, which are on the screen and are clickable. And you can also download the PowerPoint and have this information as well. We are going to be moving into a question and answer period. And we do have some time if there are questions or comments that anybody participating in the webinar would like to make. So I do again just want to encourage you if you do have questions or comments to enter those now into that chat room. Also I want to note that we are going to be taking questions over the phone if anyone has anything they’d like to say or maybe a question to ask George or Allan. So in order to do that, once again, the way that you put yourself in the queue for our operator is to press *1 on your telephone keypad. You’re going to be prompted to input your name – so you just say your name at the prompt – and then you’ll be in a queue so that our operator, Mike, can un-mute your phone, you can ask a question directly. So once again, the way to ask a question is to press *1 on your telephone pad, record your name at the prompt, and then we’ll ask for any questions and Mike, our operator, will un-mute your line. So once again, we’re going to be moving into the question and answer period now. I don’t see any questions or comments coming in through the chat as of yet. Let me just – I’m just going to give everybody a moment if you do want to type in a question or also ask a question over the phone. And let me jus go ahead and ask Mike, our operator, Mike, are there any questions coming in over the phone so far? OPERATOR: Yes. We do have a question from Edgar Garcia. MR. KEATING: Okay. Great. Go ahead, Edgar. EDGAR GARCIA: Can you hear me? MR. KEATING: We can. MR. GARCIA: All right. My question’s this. I noticed that they had two points in the presentation where the tornado started and the tornado ended. And for my use of the LED most of the time, you have to annotate the one point area where you want to go directly to and then you can play with it. You can draw or you can funnel, you can do things like that, but I wasn’t aware that you can put two locations in there. Is that a new option? MR. ROSS: You do it by – not by – (inaudible) – two points. You do it by selecting the – the selection tool is a line tool and then you can select a buffer of any width. So the way it works is you select the line option with a buffer. You put your cursor at the starting point and then you drag the cursor to the end point and then release it. MR. GARCIA: Yeah, I understand that. But my question was that you have information on two different areas – i.e. you can tell that how many people was displaced at the mall and how many people was displaced at the school. How do you get that information? How do you get that level of detail? MR. ROSS: Okay. Well, the mall in particular is a very small area. So the mall information is you select the point tool in OnTheMap and you select again a quarter mile radius. So then you just put your cursor over the center of the mall and click it and then OnTheMap draws a circle around that point. And then it’ll take – I selected a quarter of a mile because I was interested in a very small area. You have options for much larger sizes, you know, any radius – you have a wide range of radii that are possible. And so that was easy for the mall. For the tornado area, it’s pretty much the way I said. We just dragged the buffer area from one point to another and said our study area is a quarter mile buffer around this line. MR. DAIS: This is Tony Dais butting in. Edgar, I was baffled by that approach as well until I realized just what he did. He just selected a very small area and got an area profile for that little area. So that gave him the very, very specific information. That was really a cool innovation. It’s been there all the time. It’s like a blinding flash of the obvious. I can tell how many people work in a small area like a mall area, very, very specific level of data that we’ve never ever, ever had before. George tried to make the point a couple times that this narrow strip of 11 miles long and a half mile wide we could – across two states – we could never, with using traditional data sets from – never been able to get that level of data. Well, they also used the approach and technique of just pointing at one spot in a quarter mile area and you can get labor market information about the workers and who works there in a very small area. That is really important in a disaster. MR. KEATING: Very good. Thanks for those comments. Mike, any other questions coming in over the phone? OPERATOR: No, sir. We have no further questions. MR. KEATING: All right. Very good. Well, in that case we are going to do a couple of polling questions, I believe. So I’m just going to go ahead and bring up the first polling question. And the first question is: Do you find this information to be useful? And the answers that you have available are yes or no. So just going to give you a few seconds here to go ahead and answer that question. So if you find this information to be useful, please click the radio button next to yes; otherwise, please click the radio button next to no. So it looks like a lot of you are voting; thank you for that. While you do that we’re going to go ahead and bring up the file share window in case you didn’t get a chance to download the PowerPoint and the handout directly from Workforce3One with the link that’s in the chat. You can also download it directly from this file share window; you just click the file name and then click “save to my computer.” It’ll bring up the download option in your Internet browser and you just follow the prompts to download it to your machine. All right. A few more seconds on that poll. And it looks like people are still voting. And it looks like an overwhelming amount of people do find this information useful. So thank you for that. Now we’re going to go ahead and move on to the next polling question. And the next question is: Does your state have an approach like this? So once again the options are yes or no. So please go ahead and click the radio button next to the option that best matches your response. The question again is: Does your state have an approach like this? And the choices are yes or no. Click the button to vote. (Pause.) All right. We’ll give it a few more seconds while you think about that and choose your response. (Pause.) All right. Very good. It’s looking a little bit more split when it comes to this. Some of you have – some of you it looks like you do have a response like this and a bunch of you do not. And Allan, George and Anthony, please feel free to comment on any of these results as they’re coming in as well. (Pause.) MR. DAIS: Well, as the results come in I’m very gratified that we’re doing something important and providing useful information. I would say to the folks who didn’t have an approach like this, you have one now. As long as your state’s in OnTheMap – and that’s most states at this point – you can use this approach and you’ve got tools to perfect it. Now, I would note that we’ve got a comment or question in the presenter’s notes that I’m going to read to George and Allan and ask you to respond. Someone out there followed the techniques using a tornado that went through Atlanta a year ago. They say that the most difficult part was getting the tornado path as viewed from the NOAA weather site – to do that accurately on OnTheMap. You tried drawing the path about four times and this was the most difficult part of the procedure. Do you have any advice to make that easier? MR. ROSS: I similarly had a – there were a couple problems. One is drawing a straight line because I wanted it to be a straight line. And so there would up – what I would up doing was actually putting a ruler on my mouse pad and by trial and error finding how to actually drag the cursor from one to the other. And fortunately, if you – in OnTheMap if you do make a mistake you can just say disregard this one and reenter it. So you don’t have to go back to the beginning of your criteria; you can just do that portion. What I would like to see OnTheMap enhanced is to have a new – a function where you can draw – you can click two points on the map and let the tool draw the line for you. MR. PUTNAM: I should also like to add to Allan’s comments that the proposed enhancements for the next release of OnTheMap – which is currently planned for the fall – I understand includes a capability to put lat and long dimensions into OnTheMap and then to use that. So the next version will be getting a whole lot closer to a more precise delineation of area and geography. MR. DAIS: Okay. We have another question, “How does a state access the software?” Well, it’s as easy as going to www.census.gov. And look right in the middle of the page and you will see Local Employment Dynamics. And all of the tools, not just the OneTheMap tool – which is by far the sexiest, most innovative tool I’ve seen in a very, very long time to provide local, local level labor market information – there are other tools that provide great information as well: the quarterly workforce indicators and industry focus, all of which can help to provide that rich and robust view of local area. There’s no cost. It’s prepaid and it’s available 24/7, as I said. So that’s the best place. I would also say that the site – that this presentation today was purposely not a demonstration of the tool but the application of the tool. How do you use it? What is it good for? At a recent Census Bureau annual workshop on LED a private sector business person from the front of the room said the Local Employment Dynamics program and LED is like a thousand answers searching for questions. This tool can do lots and lots of – can provide lots and lots of intelligence for you. You just have to ask the question and see how the tool could be useful. So I think I’ve answered the question and some, which is usually what I do. MR. KEATING: Very good. Well, we’ve got one more polling question so I’m go to put it on the screen now. And the question is: Would you be willing to share? So for those of you who – I believe this question – correct me if I’m wrong, Anthony – but if those of you who have a tool, would you be willing to share it? And we do have a few minutes left even on this webinar. So if you’d like, you can tell us about it through the chat window and/or go ahead and put yourself in the queue so that you can participate over the phone. So once again, please go ahead and choose a radio button option that best fits your answer. The choices are yes or no. And we’ll give you a few seconds to go ahead and vote in the poll. So why don’t you go ahead and do that now. (Pause.) All right. Very good. Just want to check in with Mike one more time. Mike, anyone else on the line with questions over the phone? OPERATOR: No, sir. We have no questions. MR. KEATING: All right. Well, it looks like people are voting in the poll. MR. DAIS: I was just going to say that in the next few slides those of you who are willing to share who don’t want to talk about it right now can always send me an e-mail message to provide us information. We’re not going to have any time in this country where we’re not going to have a natural disaster occurring somewhere. The last few years have taught me that with our floods and fires and in all of the kinds of things that go on. We now have tools that will help us get a better idea of the impact and help our workforce system react better and with information. So in the coming slides you’ll see contact information for me, Anthony Dais, and please use that to share. We’re also going to be creating this community of practice and we’ll show you a slide on that. MR. KEATING: All right. Very good. And this is the slide that Anthony was talking about, so you have his e-mail address in case you don’t share over the phone or in the webinar today. But if you do have a tool or something to share, you have his e-mail address where you can contact him directly. And also, the Workforce Information and Economic Analysis tab on Workforce3One – you’ve got the Web site there – where there are a lot of resources that you can find out about what other people are doing right directly through Workforce3One. And I’ll be – we’ll be actually going over that in just a moment. But we’ll just stay here just for a few seconds. And just so you know, it looks like more people are willing to share than not, as far as the poll results go. So that’s really great. And if there are no other questions or comments we will be wrapping up the webinar in a few minutes. As you can see, the file share is still there. It’s in the upper left hand version of your screen. So you can still download those files if you haven’t had a chance to yet. And I think what I’ll do is just go ahead and show you what the Workforce Information and Economic Analysis page on Workforce3One looks like. So that is the screen shot that links to a bunch of different resources and tools that you can use basically to go ahead and find out what other people are using. So that is directly accessible through Workforce3One. And I’m just going to go ahead – hang on one sec. MR. DAIS: Yeah. I wanted to mention that under the Webinar tab – I didn’t click on that one – on that screen shot – but under that Webinar tab you’ll find three or four Local Employment Dynamics webinars that we’ve conducted over the last three or four years. So you can access them there or at the LED – the Census Bureau – site. All the webinars they’ve recorded with us and lots of other materials are located there to inform you and to provide you training. This webinar also will be recorded. So colleagues who really need to know about this, or staff members that you wish to assign to do this kind of work, you can send them to the webinar and they can get quickly trained. MR. KEATING: Very good. Thank you so much. And so what I’m going to do for just a minute here is just show you how to actually search for resources on Workforce3One that relate to this topic. So the first thing is, if you look at the top of the screen you’ve got the link to actually search the Workforce3One site. And the address is www.workforce3one.org/find. That link is clickable and that is also obviously provided through the PowerPoint. So how you would do this is when you get to this page you’re actually going to choose a category. If you notice, there are different super-categories. When you click that dropdown menu you’ll see several options. And the one that we are talking about today is the Workforce Information and Economic Analysis option – category. So that’d be the one that you choose. When you choose it you’ll see that it will show up then in that menu. And you can also search by keyword. So you can also put in words that are pertaining to whatever you’re looking for, if you’re looking for something a little bit more specific than a category as a whole. So once you do that, then you can click the Search button and it’ll bring up the search results. And if you look, you will see that the first result here is that Workforce Information and Economic Analysis resource page. And you can also see the individual resources listed. You can then go ahead and click through to the individual resources or whatever you find through conducting the search. And just a couple of other notes about that. If there are more than one page of results, because there’s often several pages, you can then choose which page you’d like to access to jump forward to results that are a little bit further down the list. And you also have a bunch of different search options. So the default is actually to sort by what it – the newest results. But you can also sort by relevance to your search, especially if you have keywords; or what resources or results are the most popular. So that gives you a little insight into how you can find resources on Workforce3One, including the kind of resources that – like that Workforce Information and Economic Analysis resource page that we were just referencing. All right. Well, we’ve got a few minutes left in the webinar. Just going to check in with Mike one last time. Anybody on the phone with a question or a comment? OPERATOR: We have no one at this time. MR. KEATING: Okay. Very good. Well, I am going to go ahead and wrap things up. Though before I do I just want to check with Anthony, George or Allan. Any last minute comments or closing remarks? MR. ROSS: Nothing from us, thanks. MR. DAIS: Well – Tony Dais – I always have something more. Thank you all for participating today. Again, this is our first win-win workforce innovators network community of practice webinar. The community of practice is just getting started. I’m actually going to a meeting following this session where the ground rules for the community of practice are being set and we’ll be launching that community of practice very soon. I thought there was a slide that showed at least a mock-up of our community of practice page. But if not, that’s okay. The next event scheduled for – I guess not. The next event scheduled for the community of practice is an event on April the 9th. We are going to have the state of New Jersey’s labor market director, Yustina Saleh, present on the real jobs in demand tool that New Jersey created that’s being very, very well received. It’s very new and she’s going to provide another practitioner’s hands-on, how-do-I-do-it kind of presentation on creating that tool to allow people to get a look at it, but also to determine whether that kind of an approach would be helpful to provide jobseeker and One-Stop center staff services – labor market information services. So look for the announcement. I’ll be sending it to all of you and one day soon you’ll be able to join the community of practice and be invited to many more how-to sessions. And thanks again. MR. KEATING: All right. Thank you so much, Anthony. Well, this is Brian again. I’m just going to close this out by giving you a little bit more information about Workforce3One. As I’m sure you know, Workforce3One is a tool built for you and by you, encouraging peer-to-peer learning among the various communities. And the success of Workforce3One relies on contributions from people like you. So just like we’ve been doing this webinar, we’re encouraging you to share your ideas, innovations and more with others. So we welcome suggestions for documents to share, programs to feature and any relevant news and information that you’d like to exchange with your colleagues. So in order to submit your content, you’re just going to visit the Suggest Content page that is featured on this slide. And that link is clickable and, once again, you can access that through the PowerPoint. All right. Please note again that all the webinar resources, such as the PowerPoint, the handout, and also the recording and the transcript will be available on Workforce3One and downloadable from that Web site. Like we said earlier, the PowerPoint and the handout is actually downloadable from this room up in the upper left hand side of the screen. You can also access it on the Web site itself. We have recorded the webinar today and that recording and transcript will be made available within two business days. And how you access it once it’s been posted is you click that Resources tab and then you go ahead and sort by Resource Type “webinar recording.” You’ll find the recording and the transcript in about two business days. And you can learn more and stay connected with trends and innovations by logging into Workforce3One where you’ll find, as Anthony was talking about, engaging communities of practice where you can share ideas, questions and innovations and connect with peers; learn more through live Web conferencing events that feature leaders and experts from industry and from government, just like this one; and a means of registering to be informed of news and events as they occur. So we encourage you to make note of Workforce3One. It’s a powerful new tool funded by ETA and powered by you. And you can also learn more about the workforce investment system by visiting www.careeronestop.org and by calling 877-US2-JOBS. All right. So with that I’d like to, on behalf of Workforce3One and today’s webinar presenters, I’m going to go ahead and conclude today’s session by thanking you all for your time and letting you know we look forward to seeing you on future webinars. Thank you so much, everyone. Have a great day. (END)