I have a really hard time grokking their lead-in "Do you expect...?" What does that mean exactly? Or more importantly, what does it mean to the respondents?
> "Do you expect Google to collect data about a person’s locations when a person is not using a Google platform or app?"
I have a feeling many people read that as "Do you want Google...." Saying "Do you expect" feels like a cynical "Do you think Google will do something you don't want them to do but are helpless to prevent, because they are a huge, uncaring, profit-obsessed corporation?" in which case, all questions can be answered "Yes" regardless of topic because the underlying cause is the same to the respondent.
"Are you aware that Google does X" might be a less loaded way to ask these.
I think asking "are you aware" is a different question. Most non-tech people have barely done any research on this and probably have never even looked into it.
You're sharing information and asking questions about it at the same time. Sort of a cross-contamination IMO.
Such cross-contamination, when done maliciously, is called "push-polling" and is a known tactic for shaping political discourse by taking advantage of being possibly the first to mention a topic in a person's field-of-view to paint that topic in the worst possible light.
I don't think that's what's being done here, but it's interesting to note that what you're seeing is definitely an issue regarding polls in general.
Quite honestly, most things that receivers call "push polls" are not push polls. They are honest attempts to figure out which lines of attack will resonate better with the public at large. So you try several different attacks on a relatively small sample, see what makes people respond most strongly, and then scale up the attacks with paid media, speeches etc. Yes, most of these will have to inform people of the bad stuff you are going to be attacking over, but that's because if the general public was widely aware of the attack, you wouldn't need to mention it- or at least not to pay pollsters to figure out if it is a good attack.
The hallmarks of a push poll are that it makes no attempt to gauge your response level, and that it tries to call far more people than a statistical survey would need to do. The most famous case, in recent US history, would have to be "John McCain's illegitimate Black baby" (https://www.vanityfair.com/news/2004/11/mccain200411)
Interesting use of the word "honest". If there's any agenda being pushed during a survey at all, then it's not an honest survey, statistically speaking. An honest attempt to figure out a line of attack isn't an honest poll.
So, any "survey" that is trying to sway or push opinions at all is a push poll in my book, even if they're also collecting information & using the information collected. You're trying to draw the line at one extreme, all push and no poll, but I believe the term "push poll" has long been used by many people for surveys that do both.
Since you can't influence the thing you're sampling and get an accurate sample at the same time, don't you think the line actually needs to be at 100% poll and 0% push, rather than requiring 0% polling before you can call it a push poll?
It’s an “honest” survey because the authors are using it to extract the data they are looking for (the most effective trigger phrasing to use in speeches, etc). It’s working as designed to provide useful information to the authors.
A push poll doesn’t provide information to authors. They don’t care about the results.
> A push poll doesn’t provide information to authors. They don’t care about the results.
That's not necessarily true, there are polls in the middle, where the authors attempt to both push agenda, and also collect information on whether the push is working. I'm saying that any amount of pushing constitutes a push poll; the defining characteristic is not whether the authors care about the results, but whether they're attempting to influence the results.
If a poll is really asking whether a certain line of negative attack is going to work better than another, and not attempting to influence the answer, not attempting to sway the respondent, then yes that would be an "honest" survey in some sense. However, there aren't actually many polls related to negative campaigns that do this, and it's also very difficult to do without influencing the respondent. Negative attack campaigns are rarely if ever interesting in doing careful neutral statistics. You'd have to be scientific about it and include controls as well as positive and neutral statements in addition to the negative ones.
So, I still don't buy the original claim that most polls on negative lines of attack aren't really push polls. They are push polls, but they might indeed be softer pushes than other more extreme push polls.
The difference, as so often, is scale. If you poll 500 people in a congressional district and ask them what their reaction to Candidate X's drinking problem is, that's an information gathering poll, not one intended to sway a significant number of voters. It might not be "scientific", but it is intended to gather information about broader voter sentiment to guide your campaign strategy lines, because even changing 500 voters minds isn't going to have much effect on the race. If you did asked the same questions, and even made the same decisions based on the data, but did it to 100,000 voters then it does start to look more like a hybrid or closer to the push-poll side of things.
I disagree strongly that scale has anything to do with it whether you call a poll a push poll. Scale is irrelevant. Asking what you think about a candidate's drinking problem is a push intended to communicate the idea that the candidate has a drinking problem, without regard to context and without discussion. You're right that it's not scientific. You're wrong that it's neutral information gathering. It doesn't matter how many people you ask that question. You can ask the same question to a single person and it's still a push, or more commonly called, gossip or shade.
Your mistake here is assuming that a push poll needs to be influencing a "significant" number of voters. It does not. Again, the defining characteristic of a push poll is the push. Not the number of people surveyed, not whether the pollers are collecting results, not whether they're looking for an angle of attack versus attacking. If the poll is pushing an agenda, if it's trying to cast doubt on a candidate, or trying to "inform" them of something bad or good, then it's a push poll, and I think the WP page made this pretty clear.
Several small scale push polls done in search of or in preparation for a large scale attack are exactly the kind of poll that count as both push and poll. Your example is demonstrating exactly the case where the pollers care about the result, and they're pushing at the same time. The intent is to influence the voter, and also measure how influenced the voter was. That's more of less the definition of a push poll.
I had a similar reaction as redwards510, but you bring up a good point. You could do an "are you aware" question first, then follow-up with "are you okay with" right after. Another alternative is "are you surprised to learn," but that's still a different question, as very knowledgeable people would not be surprised to learn it, but may not be okay with it.
The point of the survey is to test whether or not people are aware of what google is doing with their data, not whether or not they're okay with it.
Transparency is at the center of the argument of the morality of surveillance. There is nothing inherently immoral about collecting data from a consenting party, but it's impossible to gain consent from consumers who don't even know that they're under surveillance.
If you are attempting to test peoples existing knowledge of a situation, you don't want to offer new information before you gauge their existing knowledge.
> The point of the survey is to test whether or not people are aware of what google is doing with their data, not whether or not they're okay with it.
I'm not sure that's true. From the article itself: "DCN surveyed a nationally representative sample to find out what people expect from Google — and, as with a similar study we conducted last year about Facebook, the results were unsettling." Since the notion of "expect" is baked into the purpose, I think some manner of "okay with it" is fundamental to the survey. (Because when people don't expect something, that's usually not okay.)
The point of the survey is to test whether or not people are aware of what google is doing with their data
But this is a difficult question as well. Speaking for myself, I cannot say that I am aware of what Google is doing with my data. I fully expect them to use my data for any and all purposes they can think of, and I'm not okay with any of that -- but I'm still not aware of what they're actually doing.
I agree with this, but I also that "Do you expect" is a bad way to start the question. "Expect" can either mean what you think will happen, or what you think should happen. Given that, I wouldn't be surprised if different respondents were answering different questions. I'd say "Do you think Google is currently..."
We're sort of making up this whole online spaces thing as we go. Whether you love them or hate them, the decisions Google made are arbitrary and were done behind closed doors by a very small number of people.
Framing the question as "do you expect the online space to be like X" seems much more natural to me than "Google has already decided X", because the latter is likely to lead to far different answers. If we lead the question by defining it with the status quo, then naturally that's the default most people will want to agree with.
I don't think that enough people are aware of the fact that even when their data is anonymized and neither google nor its customers can extract an individual's identity, that that STILL represents serious problems.
We (the public) just don't have the vocabulary and historical experience to draw from that gives us enough perspective. There really was only one super high-profile instance of what this actually means with Cambridge Analytical and Facebook. Even that, however, is brushed off by many people.
There will have to be more examples of things going horribly wrong before people wise up and are able to understand and think through the implications of surveillance capitalism.
It's all just too abstract right now, I expect that will change in the not-too-distant future.
It's sort of an open secret that google's data is not nearly as anonymized as they would like you to believe. I'm only familiar with how advertising data works, and I can tell you that many of their large customers know exactly who they are targeting. It's not anonymous at all.
They are combining google tools and data along with their own in ways that often violate the TOS, but everyone does it anyway. I think some companies have backtracked on this in the past few months, but it is still quite widespread.
Google knows how the person is because there is a high probability that person is logged in to the Google account (which is why GDPR and such legislation do nothing but entrench these companies further).
Advertisers can't use this and don't have any way to extract identity. User ids have always been anonymous and are now completely removed from any data shared by Google's marketing software. The most an advertiser can do is target a list of users but there are limits on the minimum size and scope of these lists. It's incredibly hard to know a person's identity on the web unless they are logged into your service and it's only gotten harder.
OK, but what I am saying is that even if individual identities _remain_ anonymous, there's still a serious problem with the acquisition of aggregate information about a population.
Users are pumping out their data en-masse and entities are buying it from the surveillance capitalism "apparatus" provided by Google and the like.
That might be just fine if these entities are just companies that want to sell us soap and pop-tarts, but now we're seeing that there are other not-so-agreeable entities who want this data and are able to use in ways that the populace would find repulsive, including being incompatible with what they think is democracy.
In other words, the scary part of surveillance capitalism is that we "lose as a whole", not as individuals.
How did Cambridge Analytical impact people? People didn’t feel the pain so they do not care. Pain was felt in the Ashley Maddison leak — Google and Facebook has a lot more info than AM ever did, if it’s ever leaked it will be world ending for many.
Cambridge Analytica arguably had a major impact on the trajectory of the American political system. I’m sure you can come up with some people feeling real pain as a result if you think about it a little.
I think what the parent you are responding to is saying that there was no direct correlation between the action and the feeling of loss for most people. It is all abstract and there are plenty others to blame.
if every question begins "Are you aware..." then it sounds like Google is doing all those things, so people may start to change their answers because they don't want to appear they aren't knowledgeable. By saying "Do you expect..." it's possible that some of the things are things Google is doing and some aren't so you can't just change all your answers to yes to appear knowledgable.
I might answer “do you think Google does X” with ‘yes’ because I believe they do, and “do you expect Google to do X” with ‘no’ because I do not believe it would be right to do so.
“Expect of” is often used to connote a kind of standard setting.
I expect power multinational companies to behave ethically to the best of their ability. I believe they behave ethically only to the degree that it allows them the political leeway to behave without scrutiny.
The other problem is that there is are only two choices. They left out "I don't know". This means people who don't know probably picked more or less at random.
Asking someone if they are aware of something introduces its own set of biases, eg observer bias. Do you expect is more neutral as the respondent is not "clued in" to the "right" answer through the question.
The cynicism you describe would only be felt if the respondent is aware that tracking is status quo, and thinks that it's problematic. In which case, you'd get the same answer.
> I have a really hard time grokking their lead-in "Do you expect...?"
This is oriented to general to general users but as a G Suite customer you don't expect your business e-mails are used as input for machine learning since you are paying for the product.
I read "expect" in the sense of "Do you expect vehicle drivers to hit you, when you're crossing the road?" It's basically a question about what you believe is moral.
Doesn't ot still carry the information that they do it?
If you ask "Do you expect.." this question is neutral and you do not know if they do it or not.
"Do you expect" is neutral except it has an additional normative meaning. "I expect you to do X" doesn't just mean "I predict that you will do X". It can also mean "This is the thing you're supposed to do", even if I predict that you'll fail to do it. This is common at least in American english.
"Do you expect Google to do X" doesn't explicitly state which sense it's intended, and that can result in people saying "no", meaning no in the former sense, despite thinking yes in the latter sense.
At that point the pollsters may as well conduct a public service announcement. "Be aware that Google tracks and stores basically everything that they can. By the way, did you know this?"
"expect" is a technical term, as in bug reports where you wrote Observed Behavior and Expected Behavior. It is confusing/misleading for survey respondents, though.
> "Do you expect Google to collect data about a person’s locations when a person is not using a Google platform or app?"
I have a feeling many people read that as "Do you want Google...." Saying "Do you expect" feels like a cynical "Do you think Google will do something you don't want them to do but are helpless to prevent, because they are a huge, uncaring, profit-obsessed corporation?" in which case, all questions can be answered "Yes" regardless of topic because the underlying cause is the same to the respondent.
"Are you aware that Google does X" might be a less loaded way to ask these.