Procrastination
I've been putting this off, but there's a study on procrastination, published in early March, that I finally got around to looking at this week. It's a great study, in part because it illustrates best practices in statistical analysis. It also has something to tell us about how to avoid procrastinating.
The complexity trap
Sometimes complex analyses are preferable to simpler ones. But sometimes they're like a battalion of tanks trying to catch a fly: they won't catch the fly, but they'll make a lot of noise.
The study I'll be discussing relies on what might be called moderately complex statistical tests. Part of what makes it a great study is that the statistics are perfectly aligned with the rest of the study. The stats directly address the research questions, they treat the variables appropriately, and they're not larded with extra bells and whistles.
Far too many studies are grounded in complicated analyses that, in the end, don't tell us much. This creates what I call the complexity trap. Researchers are forced to use complex statistics, because that's what others in their field do. A researcher might look unsophisticated – and even find it hard to publish – if the fancy statistics aren't deployed. You might call this the research version of "keeping up with the Joneses".
Procrastination research 2023
One thing I discovered this week is that procrastination researchers don't procrastinate. Since the beginning of the year there have been at least 50 peer-reviewed articles on the topic. The breadth of coverage is striking. Not surprisingly, there's a lot of work on academic procrastination. Other topics include putting off health care, career advancement, and even managing clutter.
Sadly, the clutter study is cluttered with complex statistics that, in the end, say nothing plausible. (At the end, I'll touch on some older research helps explain why procrastination leads to clutter, and how those tendencies can be changed.) The study I'll be focusing on here concerns the topic of bedtime procrastination.
(If you're the parent of a small child, sorry…this is a study involving adults. You're already experts on bedtime procrastination anyway.)
The new study
This new study, published in the Journal of Clinical Psychology, was conducted by Rebecca Campbell and Dr. Ana Bridges at the University of Arkansas. Campbell and Bridges explored how sleep problems are related to bedtime procrastination, anxiety, and quantity of sleep.
This is an important topic. Good sleep is essential to physical and mental health, to productivity at school and work, to the safety of people in your vicinity, to longevity, and much more. There's a ton of data on these benefits.
Study hypotheses
Campbell and Bridges tested two hypotheses:
Hypothesis 1: People who are more anxious will experience more severe sleep problems, less sleep, and greater bedtime procrastination.
As you can see, Hypothesis 1 makes three separate predictions about the effects of anxiety. I'll say a little more about each one here.
First, anxiety was expected to undermine how much sleep a person gets. Most of us are familiar with this. The more anxious you feel, the harder it is to settle down at night. Physiologically speaking, your central nervous system is aroused, your muscles are tense, and so on. Cognitively speaking, your thoughts are racing, you're worrying about tomorrow, etc.
Second, anxiety was expected to worsen sleep problems. In this study, sleep problems were defined as daytime sleepiness as well as the sense of not having gotten the right amount of sleep. In a moment I'll say more about this prediction.
Third, anxiety was expected to promote bedtime procrastination, defined by the researchers as "delays in going to sleep at the desired time despite ample opportunity". Anxiety was expected to increase bedtime procrastination because people who feel anxious want to put off the unpleasantness of trying to fall asleep while they feel that way. Basically, you already know what a dismal experience it is to lay in the dark, experiencing anxious thoughts and feelings – and perhaps even becoming even more worried about not sleeping – and so you procrastinate about going to bed.
In sum, the first hypothesis predicts three separate effects of anxiety: less sleep, more severe sleep problems, and greater bedtime procrastination. The second hypothesis describes the how these effects link up with each other. Ultimately, this hypothesis explains why anxiety causes sleep problems:
Hypothesis 2: People who are more anxious consequently sleep less and procrastinate more at bedtime, and sleeping less and procrastinating at bedtime cause more severe sleep problems.
Here's a concrete example: You're anxious, so you procrastinate about going to bed because you know it won't feel good and you probably won't fall asleep right away. You end up sleeping less, because you're anxious, and because you procrastinated. As a result, you feel drowsy during the day, and, if someone asks, you would acknowledge that you didn't get enough sleep.
One strength of this study is that the hypotheses are are sensible, evidence-based, nuanced enough to be interesting, and clinically useful for any professional working with people who deal with issues such as anxiety and impaired sleep. And, the study design reflects a natural alignment between the way people function, the way each variable was measured (more on that shortly) and the way statistics were used to test for relationships among variables. (The chain of events described in Hypothesis 2 was evaluated by means of something called structural equation modeling (SEM) which, among other things, tested whether each link in the chain was properly described or not.)
Study methods
Campbell and Bridges surveyed 308 adults who visited a primary care center in 2019 and 2020 (before COVID-19 lockdowns). Age range was 19 to 75, and about two-thirds of participants were female.
For all of the main variables (anxiety, bedtime procrastination, sleep duration, sleep problems), participants were asked about the prior week or two. This is a good thing. The measures all concerned a recent time period, when peoples' experiences were fresh in their minds.
Anxiety was measured by three questions from a widely-used questionnaire asking participants to rate, on a five-point scale, how frequently they'd been bothered by feelings of tension, nervousness, etc.
Sleep time was defined as the difference between responses to "In the past week,what time did you actually go to sleep?” and, “In the past week, what time did you actually wake up?” I like this approach to measuring sleep time. For one thing, it's a self-report measure, but people aren't asked to record data every day, which could be disruptive. As lead researcher Rebecca Campbell put it in an email to me:
"It is certainly a possibility that being aware of being monitored impacts a person’s ability to sleep. It is also common for people to be inconsistent with their self-report or anxiously “clock-watch” if they know they’ll have to report times."
Another advantage of the way sleep time was measured is that no math is required. People don't have to recall bedtimes and wake times and subtract numbers. Rather, they answer simpler questions about when they went to bed and woke up. All the same, as Ms. Campbell noted in her email to me, "there's no perfect way to measure sleep". Although self-report measures should be fairly accurate, objective measures (like an actimetry device worn on the wrist) could improve accuracy by recording wakings during the night – though of course they might also affect sleep because they're constantly present. In the end, as Ms. Campbell notes, a combination of objective and subjective measures would be ideal.
Bedtime procrastination was defined as the difference between responses to "In the past week, what time did you want to go to sleep?” and, “In the past week, what time did you actually go to sleep?” So, if you wanted to sleep at 11 p.m. but you didn't actually go to sleep until 10 minutes after midnight, your bedtime procrastination score would be 70 minutes.
Some reflections on measuring procrastination
I have mixed feelings about the bedtime procrastination measure (as do the researchers).
On the one hand, the questions are straightforward, and preferable to something like "How long do you procrastinate at night, before going to bed?", which might seem simple but isn't really, because a person would have to think about when they wanted to go to bed, when they actually went to bed, and then subtract the two numbers. Meanwhile they might also be distracted by recalling what they did before going to bed.
On the other hand, the delay between when you want to go to sleep and when you actually get to sleep might not be directly caused by procrastination – perhaps a friend calls with an urgent problem, or you have a deadline the next day, or the fire alarm goes off in your building. (I once experienced all of these things on the same night.) The researchers acknowledged this potential ambiguity in their Discussion section. In addition, in response to a query from me, Ms. Campbell commented on the difficulties of measuring this variable and noted that she and Dr. Bridges "opted to keep [the question] vague because there are so many different reasons someone might put off going to bed."
In her email Ms. Campbell also included a 9-question scale that has been used in other studies but wasn't suitable for this one owing to time constraints. This scale wouldn't have necessarily been preferable. For example, it asks people rate how easily they get distracted by things when they want to go to bed. This might be a question about distractibility rather than procrastination. It also asks people to rate how often they're still doing other things when it's time to go to bed. This question might simply tell you how busy people are with occupational and/or domestic tasks. Et cetera.
In the end, what makes procrastination hard to measure is that you have to know what people want (e.g., to get to bed by 11) as well as the motives underlying whatever behaviors prevent them from getting what they want (e.g., checking social media late at night). Both sides of this equation are complicated. You may want to get to bed by 11, but that may not be much of a priority for you. If you only care a tiny bit about getting to bed at a certain time, staying up a half hour late on Instagram may not quite count as procrastination. As for motives, there's a gray area between checking your social media at night because you're procrastinating, versus doing so because you value keeping up with friends. Perhaps a little of both is motivating you.
My opinion is that the researchers did the best anyone could do to measure a variable that's inherently difficult to measure. I would've liked to see more questions, but since there were time constraints, I think they chose the right questions to ask.
Study findings
The results supported Hypothesis 1. The more anxiety people reported, the more severe their sleep problems, the less they slept, and the greater their bedtime procrastination.
The results almost completely supported Hypothesis 2. The more anxiety people reported, the more they procrastinated at bedtime, and their bedtime procrastination independently contributed to sleep problems.
The only part of Hypothesis 2 that wasn't supported was that amount of sleep didn't itself explain why anxiety is linked to sleep problems.
In sum, if you felt anxious yesterday, the reason you may feel drowsy today and tell people you haven't slept enough is not the number of hours you slept per se (since what counts as "enough" varies from person to person), but rather the anxiety you felt. Anxiety directly interfered with your sleep last night, and it caused bedtime procrastination, which had its own independent, negative impact on your sleep.
How to get better sleep
One of the important takeaways from this study is that anxiety impairs sleep in two ways:
–Anxiety impairs sleep directly, by creating thoughts and feelings that make it hard to fall asleep.
–Anxiety impairs sleep indirectly, by causing bedtime procrastination (perhaps because you don't want to lay awake feeling anxious).
What this tells me is that if you sense that anxiousness is undermining your sleep, two different kinds approaches to the problem may be helpful.
One approach would be to address the anxiety. Exercise, meditation, discussing your worries with someone, avoiding caffeine after lunch – there are lots of evidence-based coping strategies that help at least some people (see here for more details).
A separate approach would be to try to reduce bedtime procrastination. Here are some questions a cognitive-behavioral therapist might recommend you ask yourself:
–Can any desired/necessary activities you do at night be done earlier, so they're less tempting as excuses for procrastination?
–Can you accept that you're a bedtime procrastinator but, each night, try to shave a few minutes off your procrastination time? For example, can you shorten a late-evening procrastination activity and use the time for something that promotes sleep (e.g., a warm shower)?
–Can you at least persuade yourself that even though you're anxious and may not sleep well tonight, you'll still get through the day tomorrow, so there's no point in worrying?
How to reduce procrastination
Before anything else, it might help to know that procrastination seems to be a deeply-ingrained behavior. A 2015 twin study calculates the heritability at 0.46 – nearly the same as for IQ – which means that genetics plays almost as strong a role as the environment in creating variability in how much people procrastinate. People who procrastinate more may do so because of subtle differences in brain function, and a 2022 fMRI study suggests that procrastination arises from a neurally-instantiated cognitive biases to assume that whatever you procrastinate about, whether it's homework or chores or bedtime, the task will be easier – and equally rewarding – if you wait longer before beginning. And then there's the research on clutter that I alluded to. Studies have shown that indecisive people are more likely to procrastinate, experience clutter, and view the experience negatively. Decisiveness too is a relatively ingrained characteristic.
I think it's helpful to know these things, so that if you're a procrastinator (like me), and you want to change, you won't assume that you can change overnight. To the extent that you're grappling with genetic predispositions, neural processes, and personality traits, a little more time may be needed.
So, what can you do?
Well, I'm not an expert on this topic, and I put off finishing this newsletter until the last minute, so I'm going to punt and direct you to the best resource I've found on the topic. It's called solvingprocrastination.com, and it offers a clearly written, thorough review of the causes, symptoms, and effects of procrastination. This site also offers a detailed, evidence-based guide on how to stop procrastinating. Although the guide is realistic about the progress you can expect to make, in my opinion the title ("How to stop procrastinating") may set the bar a bit high, for the reasons I mentioned above. I do see it as useful guide to reducing procrastination, if not stopping it altogether.
There you go. If you think anything in this newsletter might help someone you know, feel free to forward it to them. (Do it now! Don't put it off!)
Thanks for reading!