You bump into someone you don’t know that well while you’re standing in a long checkout line. Mildly interested in the state of their well-being, you offer a seemingly innocent query, asking, “What’s new?” It’s been several months since you’ve seen this person, so it seems like something ought to have happened in the intervening time. However, all you get is a droll, “Nothing.”
Sometimes the questions you ask of another person have a more urgent quality. You’re trying to plan a weekend getaway with friends and want to know when is the best time to leave. Although most of your friends quickly offer up their availability, there’s one holdout who makes it impossible for you to confirm a date.
You might also know someone who never gives you a direct answer to a question. They force you to guess, ask a question in return (“Nothing. What’s new with you?”), or pretend they didn’t hear you. Such individuals make conversation difficult, to say the least.
In psychological research, refusal to answer questions shows up in what’s called “missing data.” An investigator sends a 40-item questionnaire to an online sample, and, although most people reach 100 percent participation, a stubborn minority provides incomplete data. Why didn’t they want to answer all the questions? Are they simply lazy, or was there something about the unanswered questions that they found offensive? Learning from the missing-data problem in research might help you understand the rationale behind the evasive people in your life. Perhaps there are better ways to ask questions of certain people than the direct request for information.
The Nature of Missing Data
Princeton University’s Rowena Gomila and Chelsey Clark (2022) note that “Attrition is the most pervasive and critical type of missingness in psychology studies” (p. 143). In one common form of attrition, people just skip questions when given the chance (i.e., when they aren’t required to answer every question). In some ways, attrition is the dirty little secret of questionnaire research, though it’s not really all that “little.” You may not be aware of this, but when researchers report the results from their work, it’s rare that they tell you how many holes there were in their data set.
Not answering questions isn’t limited, then, to interpersonal situations. In psychological research, investigators may not gather all the information they need because participants find the questions too personal and anxiety-provoking. Such questions may involve requests for sexual history, mental health symptoms, employment status, and opinions on controversial topics. It’s also possible that participants just become bored and don’t feel like completing the study. In any case, as the authors note, “Attrition is rarely innocuous” (p. 147).
It's the former situation, in which questions tap into some underlying set of worries or concerns of participants, that may be most like what happens in real-life settings. If some insight could be gained from how researchers can work around this problem to your own experiences, this could help you as you reword or rephrase questions that may be uncomfortable to the other person. You can also learn whether some people just prefer to remain silent.
How Researchers Can Handle Missing Data
Instead of using the tools at their disposal for figuring out why they have missing data, the Princeton researchers note that psychologists largely ignore these tools or use ones that are inadequate. They might toss out the entire data from that participant, leading to bias because that participant may have qualities that were important to know about, in addition to resulting in fewer data points in general. The missingness, thus, is not “random,” as many investigators assume.
THE BASICS
- Why Relationships Matter
- Take our Relationship Satisfaction Test
- Find a therapist to strengthen relationships
The situation can take on significance when, for example, a participant in a therapy study drops out because they’re getting worse or they simply don’t like the method of therapy being tested. Or an employee may refuse to answer questions about gender-based discrimination because the questions are too probing.
The majority of Gomila and Clark’s study provides concrete suggestions that researchers can use to make statistical adjustments for missing data both in questionnaire studies and in investigations such as the therapy study in which participants drop out for nonrandom reasons. What’s key in this paper is the fact that they show why not correcting for missing data can lead to faulty conclusions by researchers, a problem that appears to be particularly pervasive. You might want to keep this in mind as you read about research because, for all you know, the findings were biased by lack of attention to attrition.
Therapy Essential Reads
It’s All DBT, All the Time on “One South,” the HBO Documentary
Are You Stuck in Therapy?
The ”Missing Data” in Your Life
Apart from the value of learning how to be a critical consumer of research, you may now be able to figure out what “statistical” corrections you can make with those “refuse-nicks” in your life. First, when you are faced with a random non-answerer, look for cues about where your question might have missed the mark. Maybe it was too personal. Maybe “What’s new” was pretty terrible, and your question wasn’t that innocent after all.
You might also be asking the question at the wrong time. Asking everyone in a group to respond as if they were one person won’t work if someone in the group is feeling overwhelmed with other responsibilities. However, it’s also possible that, like that therapy research participant, they just don’t want to be part of the action.
The chronic refuser presents a somewhat different challenge. They may simply enjoy being a tease, or they might like the attention of watching you squirm as you try to get a straight answer. The Princeton University study didn’t look specifically at this personality trait, which appears to veer on the side of antagonism, and which could interfere with a participant’s willingness to answer. However, there are personality measures used for diagnostic purposes (the well-known MMPI) that have a built-in scale to detect people who don’t take the questionnaire seriously and answer questions in a deliberate attempt to thwart the psychologist.
To sum up, assuming your questions will be answered is a foundational rule of human interaction. You ask a question to get an answer but also to further your relationship with another person. Those people who fail to adhere to this basis for communication don’t have to sabotage your efforts to engage in the kind of pleasantries that can help promote relational fulfillment.
Facebook image: PeopleImages.com - Yuri A/Shutterstock
References
Gomila, R., & Clark, C. S. (2022). Missing data in experiments: Challenges and solutions. Psychological Methods, 27(2), 143–155. doi:10.1037/met0000361