graphs | Not To Be Trusted With Knives

Aug
10

It Wouldn’t Be Vancouver if We Weren’t Talking About the Weather

Category: Nerdery, Vancouver Tags: data, graphs, nerderyg, rain, summer, sunshine, Vancouver, weather Leave a Comment

It rained in Vancouver yesterday, so naturally everyone was all “WHEN ARE WE GOING TO GET SUMMER?????” and then I was all “Um, didn’t we have like 2 weeks of straight sunshine? Haven’t we had summer weather pretty much since April??” And then I was like this:

So then I decided to go to the data. According to vancouver.weatherstats.ca, this has been our rainfall in Vancouver for the past two weeks:

So we haven’t had *straight* sunshine for two weeks leading up to yesterday’s rainy day – we had a whooping 0.6 mm of rain on Aug 2 and 1.8 mm on Aug 3. And then no other rain in the past two weeks!

And then I looked at the temperatures. According to Accuweather, we had temperatures at or above the historical average for:

28 of 30 days in April
29 of 31 days in May
24 of 30 days in June
27 of 31 days in July
5 of the 10 days so far in August1

Here are the graphs, because all data should always be graphed2!

Now, I realize that last summer it was even hotter and much, much drier. So much drier that we were on water restrictions due to the drought were experiencing3. But this summer has been warm and sunny here in Vancouver and I’m actually sad that it’s on its way out – sunset is coming noticeably earlier and I’m having to think about whether I need to bring a sweater with me when I go out if I’m going to be out until the evening, which I haven’t had to do for quite some time. But I am glad that we’ve had a long summer and I do plan to enjoy the remaining above average temperatures we have coming for the rest of this month!

With a forecast that we’ll be at or above historical average temperatures from now until Aug 29. [↩]
Note that I didn’t make the graphs – I just took screenshots from Accuweather. [↩]
And as much as I love the heat, I prefer not having all the plants dying and not worrying if we were going to run out of water! [↩]

Jul
7

By Beth

Half Marathon #11 – A Belated Blog Posting

Category: awesomeness, friends, sports, Vancouver Tags: friends, graphs, half marathon, races, running, Scotiabank Half Marathon, Vancouver Leave a Comment

This post is ridiculously overdue. I wrote most of it on the plane to (or possibly from) San Francisco, and haven’t gotten around to posting it until just now, what with all moving and traveling and unpacking and various other goings on in my life. But better late than never, right?

In amongst the moving and traveling, I also ran my 11th half marathon, which I haven’t had time to blog about yet1, what with all the moving and traveling and running half marathons.

The half marathon in question was the Scotiabank Half Marathon in Vancouver. I’ve run this twice before – once with Alicia (2010) and once with Kim (2011). It’s a really lovely route that starts out a UBC, where you run around the campus for a bit before heading by Jericho Beach, over the Burrard Street Bridge and then finishing up in Stanley Park. The day was also lovely – it was hot, to be sure, but not insanely so, and there was a bit of cloud cover so that you weren’t running in direct sun the whole time.

The Scotiabank Vancouver Half Marathon route for 2014. Note the elevation map – specifically, the lovely net downhilledness of this race!

The one thing that wasn’t lovely was that I was sick. For the third time this spring :S I’m pretty sure that I must have picked this cold up when I was in Ontario2. I was coughing up a storm and I’m pretty sure it was sapping my energy – at least, that’s what I’m blaming my slow finish time on. I finished at 2:15:05 – more than 5 minutes longer than the Hollywood Half in April and a reversal of my trend where I’d been improving my time my last few half marathons (remember, this is finishing time, so shorter is better!):

I ran the race with a couple of friends from work. Geoff was running his second half marathon, having done the BMO half in 2013 and Christina was running her first ever half! I had photos that Christina took of us, but I think she must have texted them to me and then I lost them when my phone went insane and I had to restore it to factory when I was in Toronto. Anyway, great job, race buddies!

After the race, we headed back to UBC on the shuttle bus to get Christina’s car and then we went to Trafalgar Bistro, where we had the most delicious brunch in the history of brunch: eggs benny, but instead of being on an English muffin, it was on a potato latke. O. M. G. D. E. L. I. C. I. O. U.S.

After having done two half marathons in two months, immediately following the running study that I was in, I’m feeling like I need to change things up from just running. I think I want to find a triathlon to train for next – if anyone knows of any good ones in the Vancouver area in the fall, hit me up!

Funny that I wrote that when I typed this up on the plane – more than a week ago! [↩]
Damn you, Ontario! [↩]

Jun
13

By Beth

How Many Hours Does One Spend on Homework in a Part-time MBA Program? The Final Analysis

Category: Nerdery, School Tags: book learnin', data, graphs, MBA, Nerdery, school, stats 1 Comment

Hey, remember that time that I did an MBA? That was pretty awesome. And remember how I’m a huge nerd and was tracking all the hours that I spent on said MBA? Also pretty awesome. Anyway, the last time that I provided an updated summary, it was of the data up to the end of summer 2013. So, for completeness sake, I’ve analyzed the full data set, from the beginning of the pre-core program until I handed in my industry paper, the final thing I had to do in my MBA. (Spoiler alert: the final analysis is not much different from what I saw in the preliminary analysis). A few caveats about these data:

I completed the program in 24 instead of 28 months, so the amount of time I spent per week is higher than it would have been if I’d completed the program as scheduled
The part-time program, as scheduled, consists of a full weekend of classes roughly every third weekend, except July & August, plus work on a business plan and industry project that spanned several months and was done outside of class weekends. I took some courses from outside my program (e.g., full-time MBA classes offered in the evenings, EMBA courses offered on different weekends, some online/distributed learning courses). This makes my time spent in class less regular than it otherwise would have been.
My industry project was a project that was related to work, so some of it was done on work time. Since this was work I’d have been doing anyway, it was not counted as school work time (Though anything I did that was extra for the industry project was conducted on my own time and is counted in the data).
Transit time to and from school was not counted, unless I happened to be doing homework (e.g., reading while on the Skytrain) during that time.

And now, to the juicy data!

Overall for the two years, I spent:

831 hours on homework
641 hours in class
= 1472 hours total

If my next best alternative use of that time was to do freelance work and I charged $100/hr as a consultant, that would mean my opportunity cost for that time was $147,2001.

Overall2:

Hours spent on homework per week, on average	7.95
Hours spent in class per week, on average	6.14
Total hours spent on school per week, on average	14.23
Hours spent on homework per hour of class time	1.3

When you break it down to core3 vs. post core, you get the following:

	Core	Post-Core
Hours spent on homework per week, on average	7.95	7.96
Hours spent in class per week, on average	6.85	5.64
Total hours spent on school per week, on average	14.80	13.82
Hours spent on homework per hour of class time	1.2	1.4

The above tables tells us that:

I spent more time per week in core than in post-core
I spent the same amount of time per week doing homework in core as in post-core (regardless of the fact that I had more class time in core))
I spent more hours of homework per hour of class time in core than in core

Now, because I, along with all my classmates swore to our business statistics professor that we would always graph our data, here are some swanky graphs (click to embiggen)!

Time spent on MBA program per day

Time spent on MBA program per week

So, in conclusion, I spent a heck of a lot of time on the MBA program. However, you should note that this was my experience, but it differs from others. For example, I spoke to a classmate at graduation who told me that he spent 40 hours per week total, for the whole 28 months of the program4. And I’m sure there are students who spent less time and in different patterns, given what was going on in the lives at various points – a lot can happen in two and a half years! Basically, I’m pretty happy with my time allocation because while it was a huge amount of time and required a lot of sacrifice, I invested the time that was needed to learn what I wanted to and needed to learn, but I still managed to have some fun along the way5. But mostly, I’m happy that it’s done!

Of course, that assumes I could find 14 hours per week of freelance work for 2 years. If my next best alternative use of that time was to sit around eating Doritos, the opportunity cost would be significantly less. [↩]
Note that if I’d taken the full 28 months to do the program – assuming that the total number of hours spent on the modules I would have taken would be the same as the total number of hours spent on the modules that I did taken – I would have spent an average of:
- 6.68 hours per week on homework
- 5.29 hours per week in class
- =12.26 hours per week total [↩]
- When I say “core” here, I’m actually including the pre-core, which was three optional weekends immediately before the core, as well as the 10 months of the core program itself. [↩]
- Or possibly he finished a month or so early – so 40 hours per week for 26 or 27 months. I can’t quite remember for sure. [↩]
- Three half marathons and a trip to Europe come to mind! [↩]

Sep
27

By Beth

How Many Hours Does One Spend on Homework in a Part-time MBA Program? Another Update!

Category: Geekery, Nerdery, School Tags: graphs, MBA, sexy data, the data are crunchy, workload

A wise person once said “all data should always be graphed, always and forever”. And who am I to argue with this wise person?

As I’m sure you recall, I am tracking the time I’m spending on my MBA program. For science. When we last looked at the data, I had just finished the core part of the program and I was about to embark on the post-core. I had been led to believe that post-core was much more manageable than core and I was eager to explore this hypothesis. Of course, since I never do things simply, I’ve confounded the experiment by not actually following the part-time MBA program’s post-core program as it was laid out, opting instead to swap out some of the courses that I wasn’t particularly interested in with courses that were more relevant to my work. Which means that I haven’t so much tested if post-core is more manageable than core, but rather how my personally-designed post-core compares to core. To further complicate matters, I’ve actually taken significantly more credits than I would have by this point in time than if I were following the regular program. So really, I’m comparing core to my personally-designed, compressed post-core schedule. I’m sure that the peer reviewers of my highly scientific study will have a field day with this1. The most important thing about all of this, though, is that by compressing my schedule, I will be done my MBA by Christmas, instead of next May!

But I’m getting ahead of myself. We have gathered here today to examine the beautiful, beautiful data, not to celebrate the light at the end of the increasingly short tunnel2. So, looking at the data from the start of the program (Jan 2012) until the end of August 2013 (which includes all the courses I’ve completed thus far), the number are thus:

Hours spent on homework per week, on average	7.88
Hours spent in class per week, on average3	6.45
Total hours spent on school per week, on average	14.3
Hours spent on homework per hour of class time	1.22

When you break it down to core vs. post core, you get the following:

	Core	Post-Core
Hours spent on homework per week, on average	7.9	8.1
Hours spent in class per week, on average4	6.8	6.3
Total hours spent on school per week, on average	14.8	14.3
Hours spent on homework per hour of class time	1.2	1.3

While the numbers look pretty similar, remember that I’ve done more modules that I should have at this point. In fact, I’ve completed the equivalent of 15 post-core modules, whereas if I’d been following the part-time program as it has been laid out, I’d only have completed 11 modules5. That’s 36% more modules! So I think it’s fair to say if I were doing the post-core at the normal pace, it would have been much more manageable than core. I think it’s also fair to say that since I was going at core level of intensity (as measured by hours of work) for a year and a half straight, I can really see why I was so burnt out by the time I finally hit my holidays last month!

Now, of course, I’m sure you are dying to see the graphs. Because graphs are the new sexy.

First up, a graph of hours spent in class (red) and hours spent doing homework (blue), per day:

You can see that I was in class roughly every third weekend, with a big cluster at the beginning when I had pre-core classes for the first three weekends in January 2013 and then a big gap over that first summer, when we didn’t have classes, but there’s still significant amounts of blue that summer, as I spent the summer working on my business plan with my group. You can also see that I spent 13 hours in class on one day – that was during our capstone weekend, where we had basically one day to come up with a brilliant solution to a problem that a client was having, to which we’d only been introduced the night before. So we spent 13 hours working on it – and not to brag or anything, but my team did win, so it was totally worth it </bragging>. You can also see a thick cluster of classes in July 2013 – that happened because I took two back-to-back EMBA modules – 5 solid days in class!

I also graphed these data as weekly data, and I could identify some specific events (as noted on the graph):

So, as with my previous analysis, I conclude that the workload in the post-core is highly variable, just like it was in the core and variability is for suckers. Also, did I mention that I’m going to be done this program by Christmas time???

Thank the FSM that blogs don’t have peer reviewers! [↩]
But seriously, omg, I’m so freaking excited to be mere months away from having a life again!! [↩]
We have class roughly every third weekend, so this is the value when you average that out over the whole time period [↩]
We have class roughly every third weekend, so this is the value when you average that out over the whole time period [↩]
Note that one of the “modules” was our capstone weekend, where we met a “client” and were given an assignment on Friday night and had to present them with a brilliant solution to their problem by Sunday morning. It was EXHAUSTING. [↩]

Nov
16

By Beth

How Many Hours Does One Spend on Homework in a Part-time MBA Program? An Update!

Category: Nerdery, School Tags: book learnin', graphs, MBA, Nerdery, school, stats, the data are crunchy 4 Comments

Since the core part of my MBA program is complete, I figured it was time to check in on my “time spent” data. When last I looked 1, I had been spending, on average, 17.7 hrs per week on school (including both time in class and time on homework), with a 0.87 hrs of homework per hour of class. Since then, we had the summer off from classes, but I kept clocking homework time as we worked on our business plan project during the summer. Then we had some more classes (and, of course, homework) in the fall.

Now that the core is complete, the data show:

Hours spent on homework per week, on average	7.9
Hours spent in class per week, on average2	6.663
Total hours spent on school per week, on average	14.6
Hours spent on homework per hour of class time	1.2

So, on average, I’ve spent ~15 hrs per week working on my program4, which is at the low end of the estimate I was given when I spoke to some program alumni before I started the program. It’s an insane amount of time, to be sure – it equates to about 2 extra workdays per week, so it’s like I’ve been working 7 days per week for the past 44 weeks5 – but no pain, no gain, right? I ended up getting a good mark for the core6 – and more importantly, I learned a ton of stuff.

Now, no data analysis is complete without some beautiful graphs, right?

The red bars represent time spent in class, which you can see occurs roughly every third weekend, with a big cluster at the beginning when I had pre-core classes for the first three weekends in January and then a big gap over the summer, when we didn’t have classes. Blue bars represent time spent doing homework, which you can see varies a lot from day-to-day.

Since I didn’t really see a pattern in the daily data, I tried graphing it as weekly data, but didn’t see a pattern there either – though I could identifying some specific events (as noted on the graph):

Conclusion: workload for the program is highly variable. Variability, as I learned in Supply Chain Management class, is hard to manage. My stress level and rampant weight gain over the core program agree7!

Tonight we start our first two post-core modules for the MBA – Business Stats and Business Economics.

We’ve been told that the post-core is much more manageable than the core. Will this turn out to be true? Only time – and the data – will tell.

Back in July. [↩]
We have class roughly every third weekend, so this is the value when you average that out over the whole time period [↩]
Tee hee! [↩]
This is on top of my 37.5 hr workweek. Plus on top of teaching time from January-April, when I was teaching my stats class. It also doesn’t cover travel time, unless I spent said travel time doing homework – e.g., reading a textbook on the Skytrain. [↩]
On average. Weeks with class it’s more, weeks without class it’s a bit less. [↩]
The core program is counted as a single 18 credit “course”, so we only get one mark for the whole 10 months. [↩]
In fairness, after deciding to re-balance my life a bit by prioritizing physically activity – wherein I mean actually doing some rather than none and paying more attention to what I’m eating, I have actually managed to lose a few of the 15 lbs I gained since starting the MBA! [↩]

Feb
8

By Beth

I Snooze, I Lose

Category: friends, Geekery, Nerdery Tags: app, Geekery, graphs, iPhone, iPhone app, Nerdery, Sleep Cycle, Sleep Cycle iPhone app, sleeping 14 Comments

So I’ve been meaning since forever ago to write a blog posting about the Sleep Cycle iPhone app. It’s a free app1 that tracks your sleeping patterns2 and then works as an alarm to wake you up when you are in the lightest stage of sleep. But mostly I love it because I get to look at a graph every morning when I wake up! A graph that looks like this:

Or this:

Or maybe this:

Of course, what prompted me to write this blog posting today, of all days, is that Dr. Dan just blogged about this app! I didn’t even know that he used the Sleep Cycle app, and here he is scooping me! Of course, the data from our respective Sleep Cycle apps show that I spend, on average, 7 hrs 02 mins sleeping per night compared to his 5 hours 40 minutes per night and so clearly he used his extra awake time to scoop me on this blog topic3! As I suggested in my comment on his blog posting, “Beth snoozes, Beth loses… literally!”

Like Dr. Dan, I wish that this app had the option to overlay all my graphs on top of each other to see if some sort of pattern emerges. I mean, I have data from 243 nights4 and it’s pretty hard to spot patterns just by flipping through the graphs one at a time. I’d also like to see what my maximum and minimum sleep times were (or some other measure of the spread of my data); though my average across 243 nights is 7 hrs and 2 mins, I routinely sleep 5-6 hours per night during the week and 9-10 hours on the weekends.

Now speaking of sleep, that actually sounds like a good idea right about now. BETH SLEEPY!

and you know how I feel about free stuff [↩]
by using your iPhone’s accelerometer to sense your movements as you sleep [↩]
I have a vague recollection of Cath writing a post about the Sleep Cycle app eons ago, but a quick search of her blog came up empty, so now I’m thinking that perhaps she just told me, at some point, about her using [↩]
though not consecutive, because sometimes I forget to turn on the app [↩]

Aug
11

By Beth

A graphic look at hill runs

Category: Geekery, Nerdery, sports Tags: Geekery, graphs, I hate you hill runs, iPhone app, iPhone app review, RunKeeper, running Leave a Comment

I’ve mentioned before that I’m using the RunKeeper iPhone app to track my training runs. One of the things I really like about RunKeeper is that it gives you a graph of your pace and the elevation of the runs you do. And you know I’m a sucker for graphs.

This is a graph from last week’s hill run1:

The blue line represents pace, which is in mins per kilometer – that is, how many minutes it takes to run you one kilometre. It’s a little counter intuitive at first, because lower = faster (unlike speed, where higher = faster). Incidentally, I’m not sure why they have “speed” in the legend, since they don’t show speed on the graph2.

The green line represents elevation – it’s easier because higher on the graph = higher in real life. You can tell the app isn’t perfect though, because on this run I ran up and down *the exact same hill* four times, so the elevation at the 1 km mark, which was the bottom of the hill, should be at the same elevation as the other troughs in the graph3.

Also, I’m a little surprised that pace doesn’t mirror elevation – I feel like I run so much faster on the downhill than on the uphill. It’s starts to show up that way a little near the end, but I would have expected it to be more pronounced.

Also, I think the graph is missing a line. If I were the one drawing the graph, it would look like this:

Where the red line = will to live.

I hate you hill runs. I hate you long time.

I’d show you the graph of today’s hill run, but the end of it got b0rked, like this one [↩]
I think you can select “speed” instead of “pace” as your default measure in settings, but pace makes more sense to me when I am running, so I prefer it [↩]
i.e., just before 2km, ~ 2.5 km, and at the end of the run [↩]