Apps for Academia: Let's Talk Tech

A big part of training to be a scientist is training to be productive. Grad students and faculty do a lot of different things in a set amount of time, and that amount of time always stubbornly stays the same or reduces. Improving efficiency not only allows you to have more time for non-work things, but also allows you to devote valuable time to tasks that need creativity instead of efficiency. My advisor often tells us that we are too busy and need more time to sit around and think. 


As much as technology can be a rabbit-hole of wasted time, I love using tech for helping me stay on task. I thoroughly enjoy talking to other grad students and faculty about what programs, devices, and apps they use to organize their work lives, and I've cherry-picked from them extensively.

Here is a summary of the nine apps that I would be lost in academia without, organized from most essential to most expendable. As a warning, most of these are based on Microsoft Windows, Google Chrome, and Android platforms. Have additions or suggestions? Let me know.

Timesheet
Here is my number one productivity tool and secret weapon against wasted hours, and it is the simplest thing to do. I keep track of how much time I spend doing things. I've written about my experience before, and I have continued to keep track of my hours ever since. In fact, I've gotten so used to it, I can't start working on anything until I glance at my timesheet and decide "Ok, what am I doing right now?".

There are a million apps that keep track of time spent on various projects. I use this one, but they all seem similar.

Any.do
This is my checklist app of choice, and it nicely integrates across my web browser and phone. If I have any task to do, it gets quickly jotted down in Any.do and given a priority. I naturally think of tasks in the same way that they are categorized in Any.do: do this "today", "tomorrow", "upcoming (next week", or "someday". Additionally, I really like the daily reminder function of this app. Every morning when I start my work day (usually 8am), Any.do asks me to prioritize what I'm doing that day. Just a few minutes of considering what is important and what can wait can be super helpful when things get busy.

I used to just write down a to-do list on paper at my desk, but going digital allows me to jot something down anytime. What do you carry with you more often, a notebook or a phone?

Google Calendar
When I was an undergrad, I carried a planner everywhere I went. If you tried to schedule something with me, odds are I was going to forget if I didn't write it down. I'm still the same, except now I don't have to carry the planner and I can share my schedule with ease. Important meeting? "Google, remind me 15 minutes before this meeting starts." 

Mendeley + Scholarley
One of the worst parts of writing scientific manuscripts is controlling the references to other papers. There are different formatting guidelines for each journal, and managing huge libraries of PDFs can be a supremely annoying task. The time I was allotting to formatting citations and finding relevant papers was cut down significantly through the use of a citation management software (Endnote, Papers, Zotero are some examples). 

I prefer the free program Mendeley for a few reasons. One, it is free. Two, it is pretty good at finding information about my PDFs on its own. Three, it plays really nicely with a companion mobile app, Scholarley. Scholarley lets me access all of my Mendeley library on my phone, which is helpful when I'd like to read a journal paper away from the computer. Now I just find an interesting paper online and dump the PDF into a folder on my laptop. Mendeley takes that new file, fills in the details about the paper, and organizes it in another directory.

Maybe the most helpful aspect of Mendeley is using the citation plugin for Microsoft Word. When I'm writing and want to insert a citation, I just click Alt+M, search for topic or author keywords, and press enter. Mendeley adds the citation in whatever journal style I specify and builds a literature cited section for me. Easy.

Cloud Storage (Drive + Dropbox)
Sharing files is necessary for any level of collaboration between scientists, and the cloud storage revolution has been a welcome addition to being scientifically productive. Almost all projects I'm working on have an associated folder in Dropbox or Google Drive, where I can add/view/change content in real time. 

Pocket
Pocket is an app for your phone or internet browser that acts as a glorified bookmarks folder. I had no idea why this would be helpful, but I kept seeing it pop up on websites like lifehacker and decided to give it a try. 

The main advantage of Pocket is the ability to store all the things that I don't have time to read on the internet (blog posts, science articles, discussion boards) into a centralized place. Then when I have time, I can go through and quickly figure out what is worth reading and what's not. Very simple and surprisingly efficient. 

Twitter + Tweetdeck + Plume
I'll tell you what scientists are talking a lot about: using social media. "Uggh, do I have to tweet?" say many. No you don't, but it sure can be helpful. For scientists, Twitter can help your work reach farther, let you know what new science is being talked about, and help you connect to other scientists extremely quickly. Here is an example from my experience where I was trying to find someone who has a treadmill for salamanders. The correspondence below would have taken me at least a few emails and plenty of internet searching. Instead, in five minutes I have two prominent scientists volunteering to help little 'ol me out. How cool is that?

In terms of social media, Twitter hits the time-to-benefit ratio perfectly for me. Part of using Twitter effectively is being a little bit organized. To do this, I use a combination of Tweetdeck on my computer and Plume on my phone. I create columns for groups of people that I follow ("OSU scientists", "Herpetologists") and columns for relative hashtags ("#scicomm", hashtags from conferences). 

Spreed
Spreed is a plugin available for the Chrome internet browser that serves a very simple function. It helps me read much, much faster. Spreed works by taking the text on a page, feeding it to you one word at a time, and eliminating all the extra noise in your head when you usually read.

Now, I would never spreed a scientific journal article. I take my time doing things like looking up words and glancing back and forth at figures. However, Spreed is exceptionally useful when reading news articles, blog posts, and other internet text. 

Another potential downside is how strange you look when someones walks into your office as you are staring a words quickly flash one-by-one on the screen.

Google dictionary
Speaking of looking words up, Google Dictionary is my favorite way to look up word definitions while I'm browsing. Double click a word to see the definition. Simple.


So there you have it. Everybody's different, so what I find essential, you may find laughable. What really matters is that I get this paper done in time to watch the hockey game. 

Who Works 80 Hours a Week in Academia?

Part of being a graduate student is working hard. Whispers of disappearing faculty positions and decreases in funding percentages are heard at most social gatherings. You have to be in the lab 80 hours per week to stay at the front. Right? In this really good blog post, Dr. Meghan Duffy (follow her on twitter here!) presents her argument for why the myth of the 80 hours/week = success equation is pretty silly. 

When I read Dr. Duffy's post a few weeks ago, I had already been intensely thinking about the way I work for about a month. After reaching doctoral candidate status in the fall, I realized that I had a lot of irons in the fire. I also realized I had no real grip on how I spent my time. I knew I was working and I knew I felt like I was working hard, but I needed some more information about my habits. Ya know, data.

So how does a PhD student spend their time? Well, I've been diligently logging all of my work hours since January 14th. I'm no Steven Wolfram, but I put this app on my phone and set up six categories of my work: 

• General Work: a catch-all for emailing, running small errands, generally finding journal articles, etc.
• Meetings: anytime I was sitting in a meeting for the lab, the department, or other research activities went here
• Thesis Research: any activity related to thesis research
• Other Research: any research, including educational research, that doesn't directly apply to the ol' thesis
• Outreach Activities: this included both things outside of the university (giving a public talk, preparing materials for outreach events, etc.) and things within our lab (running undergraduate lab meetings)
• Teaching: any time spend preparing for teaching or preparing for teaching

There is definitely a lot of subjectivity involved when placing actions in one of these categories, but this is what I rolled with for better or worse. Bottom line: whenever I was really working (not surfing the internet) on one of these categories, I was "clocked-in".

How much am I working?

As of downloading the data on Sunday, I've averaged 45 hours/week of real work. I'm working an average of 6.43 hours per day (including weekends). If I look at when I'm working, I start my Monday-Friday work days on average at 9:04 am and head home at 6:40 pm on average. 

Here is a boxplot of the start times of different activities:

This makes pretty good sense compared to my general thoughts about my schedule. I usually have meetings during the morning or midday, handle little things and email first thing in the morning, and often use larger blocks of time in the afternoon for research in the lab or at my computer. I generally go to the gym early in the morning and take time for lunch with other grad students. 

What am I working on?

Since I am not teaching this semester and am being paid through a research assitantship, the data shows that I am (thankfully) working mostly on research projects. The teaching category isn't presented here for the same reason. 

Category: Duration (% of total across 33 days)

General Work: 47.46 hours (22.4% of total)

Meetings: 23.67 hours (11.2% of total)

Thesis Research: 81.28 hours (38.3% of total)

Other Research: 14.76 hours (7.0% of total)

Outreach Activities: 44.99 hours (21.2% of total)

Here is a bar chart showing how I spend each work day on different activities:

I love this view because it primarily demonstrates what I think is so great about being a graduate student: few days are ever the same. Sometimes I can spend the day devoted to a single project and some days I run from place to place doing very different things. For example, on January 30th I spent the whole morning getting some things in the lab organized, had a committee meeting late-morning, ate lunch with the departmental seminar speaker, went to the seminar in the afternoon, worked in the lab for the rest of the day, and then drove to Crawford County in the evening to give a public talk for the Parks District. The following Thursday I spend the majority of the day at my computer writing. Different days produce very different paces and styles of work.

Other observations:

I don't work that much on the weekends. I use the weekends as re-charge time and predominantly spend them with my wife and friends or working on hobbies. I think this is a good use of my time so that's what I do.

Some days are long days and some days are short days. Fridays are generally short days while Mondays/Thursdays are generally long days away from home. I'm not a robot that can work long hours day after day, so long days are usually followed by less work the next day. I can see the sense in this, but definitely see room for improvement. 

How efficient am I?

I wish I had better data to answer this question. Since I didn't log the amount of time I was "trying" to work, it is hard for me to know how many of those hours I was actually working on one of the above categories. 

One thing I was interested in is how long I spend doing each activity. For example, I am extremely fatigued by reading journal articles for more than an hour, but I feel like I can infinitely chip away at a seemingly endless "small things to do list".

Here is another boxplot showing the duration of each activity:

So, generally, I spend an average of about an hour on any given activity, but the duration is awfully variable. The extreme outliers shown here can be explained by the Museum of Biological Diversity Open House (Outreach) and the Ohio Biodiversity Conservation Partnership annual meeting (Meetings). 

But have these been a productive few weeks? I think so. I believe that I spent each of those ~45 hours to get things done. But my whole point here is to quantify what the daily life of a scientist is actually like without the posturing or the caricatures of overworked grad students pulling out their hair.

Haha, yep, there I am.

Besides the obvious benefit of having sweet, sweet data to play with, logging these hours had an entirely unpredicted psychological outcome. It turns out that taking the time to click on my phone and think "Ok, what am I actually going to do right now" is a great exercise in focusing my mind.

I'm now going to click "end outreach activity" and go watch Downton Abbey. See ya.

Data or Art?

"Spectrum Colors Arranged by Chance III" (1952) by Ellsworth Kelly

I stumbled upon this really neat post on Jim Davenport's blog, If We Assume: Data or Art?

Us scientists now have so many tools at our disposal to visualize data in new and interesting ways. While box and whisker plots may never go out of style, check out some of the sites below to find fascinating graphs, plots, charts, and more.

From Gong et al. (2011) PNAS

Wired magazine: Data as Art: 10 Striking Science Maps

The sub-reddit "Data is Beautiful"

This all-time classic talk from Hans Rosling, "Stats that reshape your worldview"

Information is Beautiful makes some very nice infographics that are pleasing to the eye

Data is beautiful on twitter


Do you have any other resources for finding beautiful, beautiful data visualization? Send 'em my way.