Interview with Jennifer Lansford, Research Professor in the Duke Sanford School of Public Policy and Affiliate of the Center for Child and Family Policy
Dr. Jennifer Lansford is a research expert in developmental psychology and family policy, particularly in the development of aggression and other behavior problems in youth. As a principal investigator for many longitudinal studies - resulting in more than 200 scientific papers – Dr. Lansford studied the impact of family and cultural contexts on these outcomes: how physical abuse, discipline, divorce, peer rejection or friendship affect the development of children's behavior problems and how culture influences parenting and children's response. She has consulted for international organizations, such as UNICEF, on standards related to parenting programs, and she has served in numerous national and international leadership roles, such as chairing the U.S. National Committee on Psychological Science of the National Academies of Sciences, Engineering, and Medicine.
- How did you get where you are today, professionally?
Jennifer Lansford, Ph.D.: I was an undergraduate at Duke, and I majored in Psychology and Cultural Anthropology. In my first year at Duke, I took a developmental psychology course, which I loved. The course instructor, Dr. Goldstein became a mentor who offered the opportunity to work in his research lab for the rest of my undergraduate years. This opportunity helped me connect with other mentors during my undergraduate studies. So, even from that time I had an interest in child development, and this carried over into graduate school.
I went to the University of Michigan for graduate school in developmental psychology. My dissertation work was a qualitative and quantitative study of family relationships and friendships in the United States and Japan. My co-advisor had a joint appointment at the University of Michigan and Tokyo University, which opened the opportunity to spend time in Tokyo while working on data collection. After graduate school, I had the opportunity to come back to Duke University as a postdoc at the Center for Child and Family Policy, where I was mentored by Dr. Kenneth Dodge.
- I understand that mentors played an important role in your career. Could you share with us a lesson that you learned from one of them?
Jennifer Lansford, Ph.D.: A couple of lessons come to mind. One is the importance of persistence in research. Persistence matters, both in terms of the research process itself (recruiting participants, analyzing data, etc.), but also in the publication process - the lesson of not being discouraged if a journal rejects a manuscript. Try to learn from the reviews to improve the manuscript, revise it, and resubmit it elsewhere. Another lesson is related to grant funding: don't write a grant proposal just for the sake of getting grant funding unless it's essential to carry out the work. Sometimes people are thinking that they want to write a grant without thinking strategically about what resources are really necessary to carry out the work.
“Don't spank your child but learn alternative ways of handling children’s behavior. Try to pro-actively manage a child's behavior to prevent misbehavior from occurring in the first place, but when misbehavior does occur, have some alternate strategies to use instead of yelling and hitting.”
- I read that you study the development of aggression in youth. Tell us an important thing that you would like to be learned by all parents to avoid behavioral problems when their kids reach adulthood.
Jennifer Lansford, Ph.D.: Parents and children get into coercive cycles that can lead to the long term development of aggression in children that can carry over into adolescence and adulthood. Pretend you're a parent in the grocery store with a young child and the child asks for candy at the checkout lane. The parent initially says “no,” and the child escalates the interaction and starts whining for the candy; the parent escalates the interaction by saying in a more harsh tone “no, you can't have the candy.” And now you see parent and child going back and forth in this way, with the interaction escalating each time. Eventually the child might win that interaction, by getting the parent to give in and say “okay, here's the candy” because they want to avoid this temper tantrum. If that happens, the child learns that “if I keep acting up, I'll eventually get what I want.” If the parent wins that interaction by smacking the child and then the child gives up, then the parent learns “If I just treat my child in this harsh way, I'll eventually get the child to do what I want.” In both scenarios neither the child nor the parent actually wins, because things have gotten to a bad point. So what you want to establish instead is parents who are having reasonable expectations, given their children’s age and developmental situation, and enforcing reasonable rules in a calm and loving way. Behavioral management strategies and corporal punishment, in particular, are real problems in terms of the development of children's aggressive behavior problems for a couple of reasons. One is that spanking teaches children that using aggression is an okay way to solve personal problems, because the parent is modeling aggression by spanking the child. So, the take away for parents is: don't use corporal punishment. Don't spank your child but learn alternative ways of handling children’s behavior. Try to pro-actively manage a child's behavior to prevent misbehavior from occurring in the first place, but when misbehavior does occur, have some alternate strategies to use instead of yelling and hitting.
- One of your research projects is focused on the role of culture in parenting and child adjustment. What are some important lessons that parents learn from cultures across the world?
.JL: I lead a project called Parenting across Cultures, which is a longitudinal study of children, mothers, and fathers in nine countries, quite different from each other: China, Colombia, Italy, Jordan, Kenya, Philippines, Sweden, Thailand, and the United States. We recruited that sample when the kids were eight year old. We have been conducting annual interviews with the children, mothers, and fathers for 12 years, and we've learned a lot of interesting lessons. Parents around the world have a responsibility of making their children feel loved and accepted. That is important for children's development all around the world. But the ways that parents do this differ sometimes across different cultural groups, and I think it can be helpful for parents to know this, because sometimes parents may think: “okay, if I tell my child “I love you,” if I hug and kiss my child, then he or she will know that love him or I love her. But there are a lot of other ways that parents around the world demonstrate love and affection to which children can respond really well. In some cultural contexts it is uncommon for parents to say “I love you” or to show physical affection, but they may show love and affection in other ways: by preparing food for the child in a special way, by making sure that the child's educational needs are taken care of, or by doing other sorts of family rituals that demonstrate love and affection. What parents are doing to show love and affection is less important than the child's interpretation of what the parent is doing as being loving and affectionate. Parents and children interpret each other's behavior within broader cultural contexts of norms in different cultural settings. If parents are doing something that's consistent with the norms of their cultural group they'll generally be getting more support from other parents within the cultural group for doing that, and children are more likely to interpret what their parents are doing as being accepted behavior. For example, a child may think “my parent is behaving in a way that's consistent with what my friends’ parents are doing.” Different cultural groups have different expectations about family obligations - for example of different responsibilities that children would have within the household or different kinds of respect that children would be expected to show to other family members. If parents’ expectations for children's family obligations are more aligned with the rest of the cultural group, that can lead to less friction between children and parents over those family obligations. This can be tricky in some families, and especially for immigrant families if children have different kinds of thoughts about what those obligations should be compared to what their parents’ expectations might be.
“Supportive parent-adolescent relationships pre-COVID protect mothers' and teens' mental health during the pandemic”
- What research are you conducting now, and what are your goals for future research?
JL: We are working on different projects, including the Parenting across Cultures Project that I mentioned, for which we are collecting data from the children, who are now young adults, and their mothers and fathers. In the last year we've expanded this work to have a particular focus on how families are coping with the COVID pandemic. We increased our data collection efforts so we're now assessing families more frequently, because things are changing fairly rapidly within the context of the pandemic. We're asking questions about changes in anxiety, depression, and substance use. A large percentage of people in our samples are reporting an increase in the time that they're spending with their families doing fun things, so it's not all bad things that are coming out of the pandemic either.
We have a couple papers under review using the newly collected COVID data, and they lead to some interesting conclusions. One is that supportive parent-adolescent relationships pre-COVID seem to have a kind of a protective or buffering effect on these families during the pandemic: not only on young adults but also on their mothers. So, those early adult-child outcomes led not only to more supportive, closer parent- adolescent relationships but they are also protective for mothers’ mental health, in particular during the pandemic. Another conclusion that comes out of the international work is that young adults’ perceptions of how well the government is handling the pandemic also are a protective factor in terms of reducing young adults’ depression and anxiety during the pandemic.
I also work on a couple of other projects that are long term longitudinal studies in the United States. One of them began by recruiting a sample of 585 kindergarteners and their families - they are now in their 40s. We have been asking a lot of follow up questions about how experiences during childhood, adolescence, and early adulthood predict outcomes in mid - life now. And we are able to see whether interventions that happened early in childhood through our study are continuing to have effects on adult outcomes and even outcomes into the next generation. In other words, we’re looking at whether adults who participated in interventions when they were children are parenting their own children in a different way than those participants who did not participate in an intervention when they were children.
- What are some of the common patterns of questionable research practices in your research area? Why do they typically occur?
J.L. The main problems that I see in my area of research are probably related to cherry picking and the file drawer problem. In large part, those are not things that people do with malicious intent; a lot of people probably wouldn't even think of those as being questionable research practices.
But I think these practices contribute to the replication crisis, which, in psychology, is that oftentimes published findings don't end up being replicated by independent research teams. Even if people try to reconstruct an experiment as closely as they can, sometimes original findings are not replicated, so I think that's led some scientists to being concerned that findings are not rigorous or reproducible. This has led to the open science movement where there is a push to bring more transparency to research findings and try to do things that could make findings more rigorous and reproducible.
Different journals in psychology have now moved to a tiered framework of open science ranging from pretty much not doing anything to very rigorous requirements such as you have to pre-register your hypotheses and analysis plan. Other journals may require authors to make their data and statistical code available on some publicly accessible website, so there are different steps that journals can take or not take along the whole continuum. There are pros and cons of these different levels; some of them make more sense and are easier to handle in certain areas of research than others. I'll give a couple examples. For example, on the issue of making data sources publicly available, some data sources are possible to anonymize so that you're not compromising the research participants’ identity while others not as much so. You can take identifying information out of computer files, but a lot of developmental psychology studies, for example, use videotaped interactions of parent-child interaction, which is much harder to anonymize. Pre-registering hypotheses and analysis plans tends to work better if you have a short-term lab-based study than a long-term developmental study, like those that I've been involved in. For example, one of the research projects I work on began in 1987. We have videotaped interactions and open-ended responses to interviews from that time that continue to be analyzed today, along with the longitudinal follow up data that we've continued to collect over the years. When data collection first started, the field hadn't yet developed certain constructs like the construct of relational aggression that was developed several years after the project began. So people who now want to go back and analyze the data that we had from that time to see if they can identify developmental precursors or consequences related to relational aggression can't really pre-register before data collection, which would be the highest standard of some of these open science frameworks. Researchers have various ways of trying to overcome these challenges, but they do highlight the issue that one size doesn’t fit all, and there needs to be flexibility.
- One size doesn't fit all and there needs to be flexibility…
JL: A type of questionable practice that deserves consideration is allowing the data to drive the message without carefully analyzing the ethical implications of doing so. As I study parents’ use of discipline and the effects of corporal punishment on children, I was involved recently in writing a commentary for an article that was trying to make a statistical point but ignored the ethical implication of what parents, practitioners, and the public would take away if they read the article.
The author had done some analysis that led to the conclusion that was stated in the abstract, that spanking has beneficial effects on children. What drove that specific statistical conclusion was the issue mentioned earlier about coercive cycles between parents and children: if a parent spanks the child, this child may immediately comply with whatever the parent said because the child doesn't want to be hit again. So, in the moment, spanking might seem to have the beneficial effect of stopping the child from misbehaving in the moment. But development happens over the long term; so parents who chronically use corporal punishment addressing children's misbehavior have children who, over time, behave more aggressively. So the study did not take this into account and looked at a specific incident rather than the accumulation, over time, which is how development actually happens.
The questionable research practice is whether the field should allow statistics to drive a message that's potentially harmful to children and to parents. Clinicians today would potentially lose their license if they're advising parents to use corporal punishment, which is now against the law in 61 countries. The United States is far behind the international world in terms of issues related to corporal punishment, but it has become a human rights issue that the United Nations is working on through the Sustainable Development Goals guiding the International Development Agenda through 2030.
Therefore - not carefully thinking about the human rights, ethical, and practical implications of a study’s conclusions and just reporting the numbers is another questionable research practice.
- What are some practices that you found most useful in your field and would like to recommend to counteract QRPs?
JL: In my area of research, it is important to be really transparent in publications about what was part of the original hypothesized model versus what was something done post-hoc. It's very possible to be transparent, say what was hypothesized, what worked and what did not. The key is to try to be clear in writing about where the shift happened and why.
I think another thing that can be done is to conduct a sensitivity analysis to test how well the findings hold up to different sorts of analyses or additional control variables.
Another issue would be to think and write in a meaningful way about whether a lack of replication is an artifact or something meaningful. In my area, especially studying development and culture, the question is whether findings are not replicating because the research is not rigorous or are they not replicating for a real reason related to development - that parent-child relationships work in a certain way in early childhood, but they don’t work that same way in middle childhood or adolescence; or the results are not replicating because of a cultural difference.
- When did you learn about these good research practices? Did you learn them early in your career? Where they required by your school?
JL: These issues weren't really talked about as much when I was in school. The whole open science movement hadn't really happened yet, at that point. I feel good about the training that I received from my mentors, but there wasn't training specifically about questionable research practices. Students who are being trained today are getting more of that professional development socialization around open science and the importance of avoiding questionable research practices. The professional organizations with which I work have workshops on these issues. Different journals build these good practices into their publication requirements. Top journals in my field, for example, have explicit statements that the methods section, for example, needs to include a sentence on whether the study was pre-registered, and the results sections need to clarify which parts were hypothesized or exploratory.
Your office is doing great work in this respect by providing information and helping people to recognize what questionable research practices are and how to avoid them.
I think that many times questionable research practices happen because people are so desperate to get a publication that they're not as cautious as they should be. For junior scholars, I think probably some of these questionable practices can be avoided by reassurance from mentors or more senior faculty that publications take time and that rigor and transparency are important parts of the research enterprise.
We have a responsibility to make sure that we're doing everything that we can to preserve the public trust in science by contributing to rigor, reproducibility, and transparency as much as we can.