Microsoft just released an online privacy study entitled, “Privacy in Action.” Intended to showcase the privacy settings of Microsoft’s products, its conclusions are a crude indication that people believe they have more privacy than they actually do.
Are you a good person?
In the study, Microsoft asserts the following:
- 33% of respondents consider privacy when choosing an online service
- 65% of respondents have deleted cookies from their browser
- 44% of respondents have opted out of targeted advertising
The issue with these numbers is that there is a high likelihood that they are inaccurate, and I suspect that this has to do with the fact that respondents self-reported their responses. Ask these same people if they are a good person and nearly all of them would probably say yes, although if we measured some objective criteria we might find that a percentage of them were certifiable scoundrels.
If a third of the population considered privacy when choosing an online service, Facebook (and other social networks) would be far less prevalent than they are. A study in India recently indicated that less than 10% of people read Terms of Service on these sites, so this self-reported number is probably significantly inflated (or people’s considerations of privacy are unrelated to actual privacy).
If two-thirds of people deleted cookies from their browser, cookies would be rendered quite useless. If 65% of people know how to do this, why is retargeting as effective as it is? One would think that a significant portion of cookies would be deleted before subsequent ads could be rendered. That isn’t the case.
The most fantastic number is that 44% of respondents have opted out of targeted advertising. Because of the deliberate complexity for users to opt-out of these ads, it is quite unlikely that number is accurate. Respondents may think it is, but they’re wrong. It’s no fault of the respondents (okay, it sort of is), but there are far more reliable ways to determine how people are protecting their privacy than a simple poll.
Not to nitpick, but….
One last thing about this study is the lack of diversity of respondents. 84% of respondents were Caucasian, 7% were African-American, and 8% were everything not Caucasian or African-American. Even if self-reporting wasn’t skewing the numbers, it would be a stretch to say that this is a representative study. This is especially true given the observed differences in internet and social usage between different racial and cultural groups.
Because there seems to be a cultural influence on certain internet behaviors, it’s quite possible that a more representative would have changed the story somewhat (actual US demographics are Caucasian 75%, African-American 13%. Hispanic 12.5%).
The state of privacy on the internet
We have a pretty good indication of the state of privacy on the internet. Google, Facebook, Twitter, and many other sites track user behavior online. Add to that list: Microsoft (through Internet Explorer and Bing). There is good evidence that very few concerned citizens read terms of service, and the only recourse for users who disagree with these is not to use any particular service.
So, is Microsoft’s study helpful to understand how people guard their privacy online? Probably not. But it does serve as a reminder that a large percentage of (predominantly Caucasian) people believe that they have far more privacy protections than they do.
What do you think – are people more aware of their privacy online than I credit them with? If this study were more diverse would the perceptions of privacy differ?