When reports say that people are more productive under different conditions, what’s actually being measured to prove it?
Someone questioned one of my blog posts the other day after he read the headline (but not the article) Are We More Productive in the Morning? He said this idea of being “more productive” that he’s always reading about sounded pretty subjective.
It’s not. It’s based on hard data.
Productivity studies, which I rely on in my work, actually do measure outputs and outcomes. “Being more productive” is not a state of mind. It’s a result that’s quantifiable.
What that output or outcome is changes based on the study. But here are a few examples.
One of the most straightforward measurements of productivity is number of tasks completed. Tasks can be taken from real world work or they can be manufactured in a lab.
Measuring tasks completed is dead-simple in manufacturing environments. It’s straightforward to measure how many widgets came off an assembly line. In knowledge work, it can be harder to find an outcome to measure, but it’s not impossible by any stretch.
A very clear example comes from Lee et al. (2012), who were investigating how rainy days and bad weather affect productivity.
The researchers combed through two a half years’ worth of records from 111 employees of a Japanese bank that processes mortgage loans. The measurement of productivity was a data entry task. When processing a loan, the bankers have to copy information from one form into another. Each loan requires more than a dozen data entry fields. It gave them 598,393 transactions to analyze. The researchers then looked at the number of loans processed on each day and pulled information from old weather reports to see if any patterns in the number of transactions completed correlated to certain weather conditions.
So long as researchers can find a task with good constants, such as processing loans the same way, every day, without worry that the number of loans that needs to be processed will hit zero, then they have a very good and clear measurement for productivity.
Lee’s study used existing data from a bank, but many researchers simply create conditions in which subjects carry out a concrete task to gauge their productivity. First they’ll find out how many of these tasks people tend to complete on average, and then they’ll change some conditions to see whether that average rises, drops, or stays the same.
In a study of how treadmill desks affect productivity, for example, subjects complete a simple typing task (Larson 2015). Researchers had some people complete a timed typing task while on a treadmill, some while seated, and some while standing. They can then compare the final results, how long it took to complete the typing task and how many errors there were, against one another to see whether walking or whether not being seated affects their productivity.
A measurement of “tasks completed” is one of the clearest ways to think about productivity. People either finish more or fewer tasks and depending on the tasks, they might do so with more or fewer errors or in a longer or shorter period of time. But it’s not the only measurement used in productivity research by a long shot.
Memory and Focus
Another common measurement of productivity is memory and focus, as determined by a test. Scores on memory and focus tests are proxy measurements for productivity rather than being direct. Researchers who use results of memory and focus tests have to (or should) first make it clear that there is a relationship between being increased focused or memory and increased productivity.
In the social science research, people often use tests developed and used by other researchers because they’ve already been proven to work. Plus, when researchers use the exact same test over and over again, they can sometimes compare their findings to those from other work, which gives them a larger body of data to analyze.
Money earned can sometimes be a fairly clear measurement of productivity in certain kinds of businesses, but it isn’t always. My work focuses on personal productivity, mostly of knowledge workers, where money usually isn’t a clear indicator. Plus, there are so many other factors that affect how much money an organization or person earns.
But occasionally money works very well as a measure of productivity that can be applied to individuals. Think of taxi drivers, for example. Taxi drivers earn money based on how many fares they pick up, how consistently they get fares in a set period of time, and how long they choose to work. A study by Camerer (1997) assessed taxi driver earnings to consider how they push themselves to be more productivity in certain circumstances but not others. The research found that taxi drivers often act counter to their best interest. They tend to quit work early under the wrong circumstances, but work later and harder when it doesn’t pay off. When they are having a strong day, making a lot of money, they tend to quit earlier. Most of them stop working once they hit an amount of money that they would like to earn for the day. But on days when they are not getting as many riders, they work longer, hoping to reach that daily minimum amount of money they want to earn.
If they thought more about their monthly wage rather than a daily wage, the taxi drivers might realize it pays to keep working longer when fares are good, and it probably isn’t worthwhile to keep working when fares are low. Time spent working on a low-income day doesn’t change the pace of the day and doesn’t significantly increase the total amount of money earned. Imagine a taxi driver is earning an average of $35 per hour on a good day, but only $15 per hour on a bad day. It makes more sense to work longer hours on the good day, as the total payoff will be higher. If a taxi driver is going to have a day when s/he goes home early, it might as well be on a low-paying day, when s/he isn’t giving up as much money.
I hope some of these examples highlight the fact that productivity research really does measure and analyze quantifiable data. The work I do is meant to take some of those data and make them more relatable and easier for individuals to apply to their own lives, habits, and work.
Camerer, C.F., Babcock, L., Loewenstein, G., and Thaler, R. H. (1997). Labor Supply of New York City Cab Drivers: One Day at a Time, The Quarterly Journal of Economics, 112 (2), 407–441.
Larson, M. J., Le Cheminant, J. D, Hill, K., et al. (2015). Cognitive and Typing Outcomes Measured Simultaneously with Slow Treadmill Walking or Sitting: Implications for Treadmill Desks. PLoS ONE 10(4): e0121309. DOI:10.1371/journal.pone.0121309
Lee, J. J., Gino, F., & Staats, B. R. (2012). Rainmakers: Why bad weather means good productivity. Journal of Applied Psychology, 99(3), 504-513. Retrieved Aug. 14, 2015 from http://www.hbs.edu/faculty/Publication%20Files/13-005.pdf
Narayanan, S., Balasubramanian, S., & Swaminathan, J. M. (2009). A matter of balance: Specialization, task variety, and individual learning in a software maintenance environment. Management Science 55(11): 1861–1876.