(no subject)
Yesterday was the City of Seattle Homeless Needs Assessment, an enormous effort bringing together the expertise of the University of Washington, City of Seattle, United Way of King County, Committee to End Homelessness, and over 500 volunteers who joined forces to gain information concerning the services that homeless people use/need and how they learn about these services, as well as the experiences that individuals who are homeless in Seattle face on a daily basis. Method: survey every single person on the street between 9pm-12am (max end time), and if they are unsheltered for the night give them a 15 minute survey and a $5 Starbucks gift certificate. As a volunteer "Team Leader" for a team comprised of myself and three other volunteers, my primary responsibilities were as follows:
As we walked through the neighborhood looking for people to survey, I often found myself considering the methodology of the study. It seems obvious that the study was meticulously engineered by individuals with considerable experience in this field. Nevertheless, I was struck by the many opportunities for inaccurate reporting that existed in this study. For example, the mandatory 45 minute training session did not even mention ways that teams should cover their study area in a thorough, methodical manner. Some teams were not even sure where their study area started and ended; when I returned to the Field Office at the end of the night I overheard a team reporting their results only to learn that they covered the incorrect geographical area and had thus overlapped with another team. Another concern I have about the accuracy of the study results is the way in which the walking speed of a team influences the number of participants they are able to study. While many unsheltered homeless individuals may remain in the same place all night, others are likely mobile. Consequently, if a team walked through their designated area slowly, they had a much greater chance of encountering potential survey takers, simply because the team was outside for a longer period of time than a team who walked quickly and completed their area in a brief amount of time. Additionally, if a homeless person walked more than 5 blocks last, they had a good chance of encountering more than one survey team. Having already earned one Starbucks gift card for participating in a survey, I see no reason why a survey taker would not feel inclined to participate in another survey in order to obtain another gift card. These are just a few aspects of the survey methodology I found myself questioning throughout the night.
Questionable survey methods aside, I think the Homeless Needs Assessment is a wonderful idea and I enjoyed participating in the study. In the 90 minutes it took my team to survey our largely residential area we encountered only six individuals. Of these, all reported having shelter for the night. Of these six encounters, only one was notable.....it went something like this:
Alece: Hi, my name is Alece and we’re conducting a survey for the City of Seattle in order to better learn the needs of the unsheltered homeless population. I wa-
Mr. Crankypants: I’m not homeless!
Alece: Well okay, we’re just surveying everyone we meet tonight and if I could just ask you one quick question: Will you be sleeping indoors tonight?
Mr. Crankypants: Of course
Alece: Thank you for your time, sir.
Mr. Crankypants: [yelling over shoulder as walking away] You aren’t going to find any homeless people in this neighborhood you know!
My word processed transcription does little to convey the vehemence with which Mr. Crankypants responded; it was obvious from his tone that he resented being asked about his sleeping arrangements for the night and thought we were foolish for asking about homeless people in his upper middle class neighborhood. My team and I made the requisite jokes about this individual going home, telling his wife about the incident, and receiving a response somewhere along the lines of, “I TOLD you to throw that ratty old sweatshirt away! It has coffee stains all over it!”.
All kidding aside, I was slightly taken aback at the man’s naiveté. Living a few blocks from our designated study area, I know full well that there are in fact homeless people in that neighborhood. Is Mr. Crankypants in denial or just oblivious? Regardless, his response indicated a strong degree of defensiveness at the indignity of (in his erroneous interpretation of my advance) being mistaken for a homeless person, and his inability to conceive of living among homeless people.
So, while it would have been more "exciting" to survey homeless people, it's important to remember that the fact that there were NOT any homeless people in that area is just as important data-wise as if there had been 100.
- Ensure that the study protocols were followed
- Introduce myself to all individuals encountered in our designated survey area, briefly explaining - the survey, and asking the qualifying question, "Will you be sleeping indoors tonight"?
- Call the Field Office to check in when my team completed half of our designated area
- Ensure the safety of my team members (aka bust out killer moves if needed)
As we walked through the neighborhood looking for people to survey, I often found myself considering the methodology of the study. It seems obvious that the study was meticulously engineered by individuals with considerable experience in this field. Nevertheless, I was struck by the many opportunities for inaccurate reporting that existed in this study. For example, the mandatory 45 minute training session did not even mention ways that teams should cover their study area in a thorough, methodical manner. Some teams were not even sure where their study area started and ended; when I returned to the Field Office at the end of the night I overheard a team reporting their results only to learn that they covered the incorrect geographical area and had thus overlapped with another team. Another concern I have about the accuracy of the study results is the way in which the walking speed of a team influences the number of participants they are able to study. While many unsheltered homeless individuals may remain in the same place all night, others are likely mobile. Consequently, if a team walked through their designated area slowly, they had a much greater chance of encountering potential survey takers, simply because the team was outside for a longer period of time than a team who walked quickly and completed their area in a brief amount of time. Additionally, if a homeless person walked more than 5 blocks last, they had a good chance of encountering more than one survey team. Having already earned one Starbucks gift card for participating in a survey, I see no reason why a survey taker would not feel inclined to participate in another survey in order to obtain another gift card. These are just a few aspects of the survey methodology I found myself questioning throughout the night.
Questionable survey methods aside, I think the Homeless Needs Assessment is a wonderful idea and I enjoyed participating in the study. In the 90 minutes it took my team to survey our largely residential area we encountered only six individuals. Of these, all reported having shelter for the night. Of these six encounters, only one was notable.....it went something like this:
Alece: Hi, my name is Alece and we’re conducting a survey for the City of Seattle in order to better learn the needs of the unsheltered homeless population. I wa-
Mr. Crankypants: I’m not homeless!
Alece: Well okay, we’re just surveying everyone we meet tonight and if I could just ask you one quick question: Will you be sleeping indoors tonight?
Mr. Crankypants: Of course
Alece: Thank you for your time, sir.
Mr. Crankypants: [yelling over shoulder as walking away] You aren’t going to find any homeless people in this neighborhood you know!
My word processed transcription does little to convey the vehemence with which Mr. Crankypants responded; it was obvious from his tone that he resented being asked about his sleeping arrangements for the night and thought we were foolish for asking about homeless people in his upper middle class neighborhood. My team and I made the requisite jokes about this individual going home, telling his wife about the incident, and receiving a response somewhere along the lines of, “I TOLD you to throw that ratty old sweatshirt away! It has coffee stains all over it!”.
All kidding aside, I was slightly taken aback at the man’s naiveté. Living a few blocks from our designated study area, I know full well that there are in fact homeless people in that neighborhood. Is Mr. Crankypants in denial or just oblivious? Regardless, his response indicated a strong degree of defensiveness at the indignity of (in his erroneous interpretation of my advance) being mistaken for a homeless person, and his inability to conceive of living among homeless people.
So, while it would have been more "exciting" to survey homeless people, it's important to remember that the fact that there were NOT any homeless people in that area is just as important data-wise as if there had been 100.