Never noticed a difference in this but I have noticed a difference based on where you were trained/background.
What I always do first thing rather than diving right in to explore and play around, is investigate the data in a database/tabular format (lame, I know, but it's just how my brain works). I determine, what is a row of data in this source? That will guide me throughout the discovery process, and knowing the structure really well usually means I can avoid the pitfall of "hey I found this amazing finding but wait isn't there supposed to be only one row per ID?" I have learned this from years of not doing this first, and then realizing that not knowing the data was undermining my conclusions.
I think it's tough to do this with Tableau, because the products make it so easy to explore. Other products I can think of, statistical packages for example, have a much higher bar of entry. But a good thing about them is that often the first thing you learn how to do is to compute an average or variance or other summary statistics, which does force you to understand the data before you continue on to try and draw conclusions from it.
Thanks for sharing Erin!
That's a really great point. I definitely fall into the category of wanting to explore the data right away. When I'm presented with a new data set, I usually pull the different dimensions and measures up onto the view to try to understand what each one is showing. Then, once I understand what type of data is included, I start asking the questions. Sometimes asking I have a question in mind before I even know what's in the data - so I'll make guesses about what different fields might tell me. It definitely can be a bit of trial and error.
Can't say that I've really ever noticed any difference between men and women in how they approach the data analysis either. I think a lot of how people approach the data depends on what they know about it and what their roles are. For example, Patrick Van Der Hyde and I often look at the Community data, but he's on the Support side of the house, and I'm on the Marketing side, so we have different perspectives on what we're looking for.
Yes as a Male I totally agree with you...
Thanks for sharing Erin!
That's a really great point. I definitely fall into the category of wanting to explore the data right away. When I'm presented with a new data set, I usually pull the different dimensions and measures up into the view to try to understand what each one is showing. Then, once I understand what type of data is included, I start asking the questions. Sometimes I have a question in mind before I even know what's in the data - so I'll make guesses about what different fields might tell me. It definitely can be a bit of trial and error.
I can say that I've really never noticed any difference between men and women in how they approach data analysis -- but then I'm a male. I think a lot of how people (male or female) approach the data depends on what they know about it and what their roles are -- and what their 'assignments' are. Or maybe I simply agree with Erin that there is little difference (based on sexual organs) as to how we individuals think about data. Or approach data.
Tracy why are you attempting to draw a line (distinction) where there is none?
We are simply people. We each approach data analysis in our own unique (individual) way. Brit has her way. Kelly has her way. I have my way. And you have your way. Do you really think our Gender will define these different individual's approach to data? Or do you simply need to promote Women + Data on our forums?
Do all women think the same? Are all women alike? Do all women have the same life experiences?
The answer to all these questions are the same for 'All' men.
I'm not drawing a line (or attempting to). I'm starting a conversation.
I didn't suggest that there was a difference. I'm honestly curious as to whether other people feel that there is - and who knows? Maybe others have noticed a difference. As I said, I haven't, but the only way to learn is if people talk about it.
Thank you for providing your perspective.
But this is not to say that gender does not influence where and how we get our training, at all. We all know that women are underrepresented in science and math, and so it's very likely that a woman who is a data analyst might have become a data analyst not through formal training, but through other experiences and jobs, which of course will influence how she approaches data analysis. Projects that one is assigned in school and at work often reveal biases that are then reinforced throughout ones career (Amy-you clean the data, Anthony-you build the model), which would of course influence how someone approaches data. Anthony would be pretty comfortable just diving in and going for high-level conclusions because he never had to clean the data. Amy would be hesitant to even start before the data was clean.
Fair enough. You started a conversation. It seems the crux of your conversations-starter is:
Have you ever seen any differences in how men and women approach data analysis? And...
What can women learn from men in how they approach data? What can men learn from women?
Your assumption in both these questions, 'conversation starters' is that men and women are somehow different in their thinking. How? Is your (female) mind/thinking different because your body contains estrogen? Is your husband's (males) mind/thinking different because his body contains testosterone?
Logically the answer is: No. But that is not the end of the story. Women are different from men. You simply have to decide how you want to make that distinction. Or from my point of view: 'Fairly' make that distinction.
My most fervent wish is that ALL countries, states and municipalities were governed by women throughout the world. Peace would eventually reign.
PS: This is probably the most sexist comment I will ever make in my life; but I sincerely wish this were true.
You want parity. Stop thinking 'Men' & 'Women'. Start thinking we are simply people.
Look at the pay discrepancy and we'll start there Shawn
Notice how the top ones are "softer" careers. Women tend to be "softer" so some gravitate towards those fields, would be my guess.
As far as the original question, I feel like my analysis always leads me to more questions. I always start with a question, try to answer it and show the best answer via story telling in tableau, but I typically uncover MORE questions I need answering. I don't know if men's minds work like this, too? My mentor is also a woman and her mind is very similar to mine. We both are very curious in our analyses which leads to further depth. I don't just see the problem "I need to see what percentage of workorders were done on a standard job per store" I take that and run with it. I cant speak for my male counter parts because I'm the only one with my role in my company.
In my Tableau life, there are two kinds of people, who are addicted to cross-table figures and who are open mind for creative visualizations
This makes difference
The way I present data tends to be filtered through a male lens (of course, along with the many other lenses that I inhabit).
But let's look at this viz for a second (also a shameless plug for the vizzes I build)
I remember finding this data set, building out the view, and publishing the viz with a sense of detachment.
The data tells me a story of deep injustice and the inaccuracy of many ingrained narratives. Yet I am unable to truly comprehend what it means to face those odds or know where I fit in.
With regards to how I tell the story--even if it isn't related to gender wage gap data--I tend to disconnect from the meaning when it steps outside my realm of experience.
I personally feel that my presentation has the following characteristics:
- Does not encourage discussion as much I'd like
- Merges fact with conjecture too much
- Is too divorced from emotional impact
Now, some of that is simply the immaturity of my time in the saddle. But I'm certain a lot of it is also characteristic of my maleness.
I know that when it comes to storytelling, I often look to people such as Jewel Loree
When I want to make a unique viz-type I look for tutorials from people like Jeffrey Shaffer
I feel that tutorials, from an unscientific sample, skew male while vizzes that I remember the most skew female. Is that a pattern anyone else notices? Is that pattern real?
Also, please challenge me on any assumptions I have made.