Even though this is my fifth year at Augustana, the concept of Muesday still throws me for a loop. Maybe this is because I don’t have to think about it much, counting beans in my little office all day every day like I do. Conversely, most faculty I know talk about it as if it’s the most normal concept in the world, no matter if they’ve taught at Augustana for a couple of years or a couple of decades. And even though I think I’ve developed a failsafe cover to hide my ignorance (toss my head back, laugh, lean in while I bat the air in front of my face, say emphatically, “of course, what was I thinking!” while rolling my eyes), it’s an annual reminder for me that the concepts each of us take for granted aren’t always so obvious to everyone else.
I’ve been reminded of this reality again as I’ve been inviting everyone to fill out current Augustana College Employee Survey. More than a few people have expressed concerns about anonymity and confidentiality. A few have even floated impressive conspiracy theories of NSA-caliber data scrubbing. So before I have to run off to my weekly administrator neural network reprogramming and empathy reduction session, I thought that I’d try to answer the anonymity and confidentiality questions in a little more detail. (Yes, I’m kidding. The administrator neural network reprogramming and empathy reduction sessions are every OTHER week and don’t meet this week because it’s MUESDAY!)
When I promise anonymity to everyone who responses to the Augustana College Employee Survey, that means that I don’t ask for your name or other information that directly identifies you. It also means that the software doesn’t collect your Augustana user ID or the IP address of the computer that you used to complete the survey. In order to do this, I turn off a setting in the Google Forms software that would normally add this information to the dataset.
Turning this feature off also means that the survey is publicly accessible – a potential downside to be sure. So it is technically possible that each of the 340+ survey responses I’ve received aren’t actually coming from Augustana employees. But that would mean that somebody somewhere else has acquired the web address of the survey and has spent their days and nights repeatedly filling the survey out over and over with just enough variation of answer choices to avoid suspicion. Yeah, I doubt it.
Some folks have pointed out that there are enough demographic questions that there might be a way to identify some respondents. This is technically true: if someone had access to both the college’s employee database and the current employee survey dataset, one could probably figure out a way to be pretty sure about the identify of some of the respondents, particularly if one were to triangulate several demographic characteristics (e.g., race and age data) to pick out subgroups of employees that have only a few members. Of course, the only person on campus who has access to both of these datasets is, well, me. If you think that this is a likely explanation for how I spend my time … I guess I sort of doubt that you are even reading this post. Nonetheless, to be clear – I’m not trying to figure out what you said in your survey. And I’m not taking that information and slipping it under someone else’s door so that they can hire henchmen to come to your office and hide your keys. It’s not that I don’t care. I’m just too busy.
All joking aside, this survey does ask some questions that can easily be perceived as risky to answer. So, if you are concerned about anonymity but want to respond to the survey, just leave any demographic question that cuts too close to the quick blank. That way you don’t have to worry about having your anonymity violated. I think we’d rather be able to stir your opinion into the mix even if it might not get included in more complex analysis.
Confidentiality is a little different from anonymity. There are numerous student surveys where we promise confidentiality but not anonymity. We often will ask students for their ID number so that we can merge the data they provide with prior institutional data so that we can take a longer view of our students’ four years at Augie, looking for patterns across the entirety of their college experience. Confidentiality specifically refers to how we will share any of our survey findings. When I promise confidentiality, I am promising that I won’t share the data in any way that might link your set of responses to you. Instead, all data findings will be shared as averages of groups, whether that be the entire group of respondents or small subgroups of respondents.
This does again raise the question that some have asked about protecting the anonymity and confidentiality of those who are members of sparsely populated subgroups. When I promise confidentiality, I have to also consider the possibility that presenting data in all of the ways that it can be sliced and diced could lead to violating someone’s confidentiality. To allay this concern, I am ensuring confidentiality by simply not sharing any results in a way that might allow folks to reasonably infer any individual’s responses. I will not share any average responses to questions where the number of respondents in that particular subgroup is less than five. This makes it much less likely that anyone could determine the nature of someone’s individual responses based on the average responses from any particular subgroup of responses. So, for example, if we have less than five respondents in the category of employees who have worked here between six and ten years, then we won’t share any results for any question by the number of years employees have worked at Augustana.
Just like the anonymity question above, if you are worried that your confidentiality will be violated, don’t provide answers to those specific questions.
Even though we have already received many responses to this survey, we still need many more because the more that we have, the more likely it will be that we can look at subgroups of responses and analyze this data without violating anonymity and confidentiality.
Getting better as an organization is hard work. At its core, it requires that we all put something into it. Completing this survey is a big first step. I hope you’ll all give it a shot.
Make it a good day,