OTLA Trial Lawyer Fall 2020

48 Trial Lawyer • Fall 2020 Mechanical Turk and surveys The original Mechanical Turk was built in 1770 by Wolfgang von Kem- pelen to impress the Empress of Austria. He claimed it was an “automaton” ma- chine that could beat anyone at chess. It turns out it was just a box with a chess master hiding inside moving the pieces. Amazon called its platform Mechanical Turk because it ran into trouble creating algorithms for some of its processes. It created Mechanical Turk, the website, to crowdsource these jobs to real live human beings. In recent years, the platform has become a common place for market re- search. Academics now accept the plat- form as a sufficiently representative sample of the population, and it has become a go-to-place for graduate stu- dents to conduct their research surveys. See https://daily.jstor.org/amazons-me- chanical-turk-has-reinvented-research/. In MTurk, you can customize the demographics as much as you want, but the more specific the demographics, the more expensive and the longer it will take to get results. For instance, if we have five yes or no questions we want answered by 100 people and it doesn’t matter who, we can post it onMTurk, offer to pay people $0.10 per survey, and we will get 100 replies in a matter of minutes. On the other hand, if we want 50 people to answer a survey that will take 20 minutes, who are from Oregon, female and over 40 it will cost $5 to $10 apiece and take at least a week to get. You set the price for the survey, but MTurk recommends paying approximately minimum wage. The trouble with MTurk is that con- ducting an actual survey on the platform is cumbersome and technical. MTurk is really just the platform that connects surveyors with surveyees. We couple MTurk with Survey Gizmo because it has a much easier interface and is built for collecting survey data. Other survey software work just as well (Survey Mon- key or Qualtrics). The basic process is you post the survey on MTurk. Once users log on, they click a link that takes them to the survey on Survey Gizmo. At the end of the survey, they enter their MTurk ID, which Amazon authenticates. We then approve payment for all the authenticated IDs that completed the survey. If this sounds complicated, you just need to try it once and then it be- comes fairly straightforward. We recently had a case where we wanted to know how our client came across to potential jurors in order to see if this was a case we should be focusing more on taking to trial or more on set- tling. We had our client record on her phone a 10-minute video where she told her story of what happened. Participants were able to view the video and then answer questions to help us gauge the believability of her story and help us see where the story had holes that needed to be explained. This is a great platform to really test a theory because you can get a lot of re- plies and once your sample size is big enough, you can take the majority as a consensus. It is also nice to “fact check” what a focus group is telling you. If you can ask a few questions that line up with what the focus group has stated, and it comes back with the same perception, you can be reasonably secure you have an accurate picture. A couple things to watch out for. Every survey platform has pros and cons. Surveys, unlike focus groups, do not have interaction with the participants. You can ask for participants’ emails to follow up, but it is not the same exchange you get in person. Sometimes it is difficult to craft questions that will give you reliable data. Most of the survey apps have a free trial period, so you can test them out to see what works for you. A problem we had when it came to surveys is we had no idea what we were doing. This is where Freelancer.com, Upwork.com and Fiverr.com helped. We posted a job for someone who was trained and educated in running surveys and had experience. We interviewed a few people (via Zoom) and found a Ph.D. in Experimental Psychology and Research Methods who is a good fit for us. When we have a particularly tough survey or complicated issue, we use her to help us get the survey just right so the data is reliable. This is a little more sp- endy, but it takes the guesswork out of our testing. Overall, surveys can be a very effective tool for finding general attitudes on subjects, as well as getting a large number of replies, and it’s cheap. The downside is any follow-up is hard to do, you don’t get to work on your people skills and it has the potential of being worthless if the right questions aren’t asked, but that’s true with any focus group. What do you do now? What you do is you start focus group- ing. Play around with ideas, try new things. The worse thing you can do is think you already know what regular people will think of your client’s case. One of the best ways to overcome your blind spots is to ask questions of potential jurors. Whatever you do, do something. Don’t just sit there and think “I should really do this” and then do nothing. Test- ing is a huge part of many lawyers’ success. Why guess, when you can test? Earl and Erin Christison specialize in nurs- ing home negligence cases. Together they own the firm Paynes Grey, LLC. Their firm is located at 5955 Shoreview Lane N. Ste. 102, Keizer, OR 97303. They can be reached at [email protected] and erin@ paynesgrey.law or 503-390-2044. 1 Even in writing this article we disagreed with the editor about what readers would want. After focus grouping the introductory para- graph we learned we were wrong, and went with the editor’s version, which shows the power of focus grouping. If you would like to take the survey go to this link: https://www.surveygiz- mo.com/s3/5930757/5-Minute-Survey What was telling from the survey is that al- though attorneys do not get offended easily they are very concerned with offending others. The Virtual Way Continued from p 47