Command of evidence: Quantitative — Worked example

- [Instructor] Let's take a look at this question from the reading and writing test. So right off the bat from the table here, I know I'm looking at a quantitative evidence question. We have an infographic full of data, a short passage that outlines a trend in that data, and we're asked to figure out which choice most effectively uses data from the table to complete the example. But before we get to deep into this particular set of data, let's talk about quantitative evidence questions in general. So if you'd like to try this item on your own, feel free to pause the video now and give it a shot. We'll be returning to it and working through it step by step later in the video. All right, let's move on. So quantitative evidence questions require us to combine our reading comprehension skills with our ability to read and interpret infographics, like graphs or tables. These items will appear a few times on your test. Your goal in these questions is to select data from the graph or table that effectively supports or exemplifies the argument being made by the short passage. So essentially, you're given a haystack of data and your job is to find the needle, right? To find it, you need to know what sort of needle, what sort of data you're looking for. So your first task for these questions is to understand the argument being made in the prompt. We can't know what data to hunt for if we don't know what's being claimed. Here's an effective strategy for working through these questions. First, skim the graph or table, familiarize yourself with what's being measured but don't linger too long. You wanna get a sense of the units of measurement, any labels or titles, and any key or legend. We won't know what data to look for until we understand the passage. So the next thing you need to do is identify the argument the passage is making. This should tell you what data you'll be looking for. Once you've figured out the argument, boil it down to a test phrase of a few words and then test that test phrase against the answer choices. Only one choice will match your test phrase and support the passage's argument. This is feeling a little abstract to me. So let's put this into practice, and you'll see what I'm talking about. Let's go back to that example question. Step one, let's quickly look at the table here: Participants' Evaluation of the Likelihood That Robots Can Work Effectively in Different Occupations. Okay, so the study is asking people, do you think a robot can do it? And then giving them several different human occupations, TV news anchor, teacher, firefighter, surgeon, tour guide. So we got a couple different jobs in the left column. The top row shows us the possible survey responses, unlikely, neutral, and likely. And that's it. We can move on. That's the level of depth you're looking for, 10 seconds worth of investigation. I don't wanna waste time by digging in to deep yet. So let's move on to step two and identify the argument. So I'll read the passage. Georgia Tech roboticists De'Aira Bryant and Ayanna Howard, along with ethicist Jason Borenstein, were interested in people's perceptions of robots' competence. They recruited participants and asked them how likely they think it is that a robot could do the work required in various occupations. Participants' evaluations varied widely, depending on which occupation was being considered, for example, blank. I think that last sentence is the key. Participants' evaluations varied widely, depending on which occupation was being considered, for example, blank. So what does that example need to show? It needs to show that evaluations varied widely. This is the key idea we need to provide evidence for. So we need data that shows how people thought there were some jobs robots could do well and other jobs that robots can't do well at all. So my test phrase is something like a wide swing depending on the job. That's what we're looking for. So let's test a wide swing against the choices. Choice A, 47% of participants found it somewhat or very likely that a robot could work as a teacher. That's in the table. 37% found it somewhat or very unlikely. That's also in the table. So this is a 10% swing, and it only talks about robots having one kind of job. This isn't it. Choice B, 9% were neutral about robot TV news anchors, which is the same percent of participants who were neutral about robot surgeons. So this compares two different robot occupations but the difference between them is zero. So this is not our choice either. Choice C, 62% of respondents don't think robot firefighters would be very good and this doesn't illustrate a range, right? And it's also only about one profession. So we can eliminate choice C, which leaves us with choice D. 82% think a robot makes a likely tour guide. 16% believe it would be very likely that a robot could be a surgeon. That's a pretty big swing, no? 82% pro tour guide versus only 16% pro surgeon. And it's the only choice that both shows a wide swing and compares to potential robot jobs. So this is our answer. Choice D compares two robot jobs and has the widest swing and is therefore the choice that best supports the argument in the passage, and we can select it with confidence. Let's go over some top tips for quantitative command of evidence questions. Find the story. Data tends to tell a story. It shows similarities, differences and changes over time. The passage will reinforce this story by drawing attention to it. For example, here I identified the story that the text was trying to tell, that there'd be a wide swing in the data when we compared two different robot jobs. And once I know what that story is, it becomes a lot easier to identify the choice that tells the same story. Be flexible. There are a few different ways for this table to express a wide swing, but we're not looking for a specific data range. We're looking for any data range that supports our argument. In our example, a wide swing could have been shown in a few ways. The answer showed how a lot of people thought robots could be tour guides but not many people thought robots could be surgeons, but there are other combinations of data from the table that could tell the same story. So if we approach the choices with a particular combination in mind, we might miss an equally effective piece of evidence among the choices. So we need to be open and flexible to all possibilities. Good luck out there, test takers. You've got this.