When No Response is a Response: Predicting Survey Non-Compliance in Isolated, Confined, Extreme Analog Space Missions

Long-duration space exploration will place small, interdependent teams in isolated, confined, and extreme (ICE) environments, where they will face risks including communication delays, high autonomy, increased workload, team tension, and intense stress. All of these risks require research studies aimed at mitigating them, many of which take place at analog space stations. However, these analog missions, which aim to study human behavior in ICE settings, suffer from chronic survey noncompliance, which threatens data integrity and reduces the accuracy of crew monitoring. This project examines which individual characteristics predict who completes research surveys during missions at the Mars Desert Research Station, a Mars-analog habitat in Utah. Before their missions, crew members completed questionnaires assessing personality (with a focus on conscientiousness), their identification with their team, their cultural values (especially collectivism, or putting the group before oneself), and their difficulties with emotion regulation. During the mission, they were asked to complete brief daily and weekly surveys. For each person, I will calculate survey compliance as the percentage of assigned surveys completed, and I will use regression models to test whether conscientiousness, emotion regulation, team identification, and collectivism predict higher compliance. I will also test whether collectivism strengthens the link between team identification and survey completion, such that highly collectivistic crew members who strongly identify with their team are especially likely to respond. I expect that crew members who are more conscientious, better at managing their emotions, more strongly identified with their team, and more collectivistic will show higher survey compliance. The work integrates literature on ICE stressors, organizational citizenship behavior (OCB), survey methodology, and individual differences to generate actionable recommendations for designing analog-station research protocols and for operational monitoring in deep-space missions.

Hi, my name is Amber Bartlett. I’m an honors psychology student at George Mason University. My project is called “When No Response Is a Response, predicting survey non-compliance in isolated, confined, extreme, analog space missions”

My mentor is Dr. Lauren Kuykendall in the Industrial Organizational Psychology Department. So what is the big picture of my project? Long duration missions to Mars and to the moon will place very small crews into what we called isolated, confined, extreme, or ICE environments. These missions will involve tight living quarters, high workloads, autonomy, communication delays, and one of the tools we can use to track how these people are doing is through self-report surveys.

But the issue with doing research on ICE teams is the fact that the sample sizes are incredibly small, so when people skip surveys, we lose statistical power and increase the risk of error in our findings, and missingness is often not random. The people who stop answering might be the ones who are most stressed out, which means that things can be biased. So this leads to my core question, can we predict who complies with surveys before a mission? If we can identify likely non-responders ahead of time, then we can actually anticipate, model, and design around that.

So this project is a part of a collaboration between the University of Central Florida, where we’ll be using UCF’s IO Psychology PhD student Andres Käosaar’s data set from the Mars Desert Research Station. So MDRS is a Mars analog habitat where small crews simulate living on Mars. They’ll be conducting EVAs, conserving resources, and living in a confined environment.

And for this project, I will be using the data set that includes 99 individuals, 16 crew members, and a total of 761 habitat days. Most of the missions had teams of about six to eight members, and participation in Andres’ study was completely voluntary. Now because it was voluntary, I treat filling out these surveys as a type of organizational censorship behavior, and this is when people are doing small extra things at work that help their team organization, because doing his research survey was completely just to help his research.

So to understand why some people keep doing this extra work and why others don’t, I focus on four key constructs from the literature. So that’s including conscientiousness, emotional regulation difficulties, team identification, and collectivism within someone’s culture. So the knowledge gap is that analog space research already acknowledges chronic survey non-compliance and missingness.

However, individual level predictors or survey compliance and ICE teams remains undermodeled. We don’t have a clear guidance on what type of people are most likely to disappear from the data set, so my project is designed to help close that gap. So within this data set, participants completed pre-mission surveys which included measures such as conscientiousness, emotional regulation difficulties, team identification, and collectivism.

So during the mission, they received daily and weekly self-report surveys about their experience and well-being, and my main outcome is survey compliance, so this is defined as the percentage of assigned surveys each person completes over the mission. So based on this, I have five different hypotheses. First is higher conscientiousness will predict greater survey compliance.

Second, better emotional regulation will predict greater survey compliance. Stronger collectivism, stronger team identification, and collectivism moderates team ID compliance. All of this, I’m going to assume it will increase greater survey compliance.

Now with OSCAR’s help, I was able to visit MDRS in person, and these photos here show the habitat. It shows the people that I was with out there, and it really helped me understand what it’s like to be living in one of these space analog stations. So my next step is to run descriptive analysis on all of the hypotheses that I just showed you, and so that will be what I’m doing next semester.

The expected implications is that if my hypotheses are supported, this work suggests that voluntary survey compliance in ICE analogs is not random. It’s a predictable citizenship-like behavior rooted in personality, emotional regulation, and how people relate to their teams and groups. So this has several implications such as pre-mission screening, low burden monitoring, and then just identifying patterns of non-response, but also if my hypotheses are not supported, it means we should be looking somewhere else other than the OCB literature, which right now is where the main source of research is coming from when it comes to non-response and surveys.

So I want to thank my mentor, Dr. Lauren Kuykendall, Andres Käosaar, Dr. Seth Kaplan, Dr. Brielmaier, my psychology honors cohort, the MDRS director, and of course OSCAR URSP. If you would like a link to all of my resources that I’ve used, here’s a QR code for that. Thank you so much for listening.