Jeylan Mortimer
Youth Development Study

Overview

Q: Could we start with you giving us just a one or two minute summary of the project?

A: The study was initiated in the late 80s, actually in the 1987-88 academic year. The study, when it was initiated, was mainly concerned with the whole issue of teenager's employment and whether youth should be working, how many hours they work, what kinds of work they do. At the time there was very little known about this whole phenomenon, though people were aware that young people were in the labor force and there had been some studies but none that were longitudinal and dealing with a community sample, followed over time, so we started off the study with three basic questions. One was, "What kinds of work do young people do and how does that work change over the course of high school?" Secondly, "What are the impacts of work on youth?" There was a lot of controversy about whether working drew adolescence away from school, whether it affected their grades, whether it harmed their relationships with their families and also with their peers, how it would affect their psychological and psychosocial development. So we were concerned with all of those questions, and then we were also interested in how working influenced the transition to adulthood and whether early work might be a source of human capital and social capital that would be useful as adolescents made the transition into full-time work and also into higher education. So these are the basic questions that we started with.

Theory

Q: What was the role of theory in this study?

A: For a long time there's been speculation and some theory about the impacts of work on psychological and social development. Some think that working is a source of fulfillment and a very positive factor in people's lives; others think that work can be otherwise and emphasizes dehumanizing aspects, how work can be oppressive. Marx, Durkheim, this extends from very early thinkers in sociology. In my earlier work, I became interested in the life course and work, and whether the question whether working might have different implications depending on the stage of life that a person was in, and this extends from life course theories, the work of Glen Elder and others, concerns about developmental readiness, the issue being if you begin work perhaps too soon, you may not be ready for it in a number of ways, and this could become a detrimental factor in development. But then the opposing view is that early experience with an area of life that is very important in adulthood would provide learning experiences and benefits in the future. So initially the study started with these broad questions that are quite substantive in nature. We were not dealing with some grand theory at that point.

Q: There've been many different people doing many different analyses and studies with this big study. What are some of the more recent theoretical questions that people are exploring, or that you are exploring?

A: Well, one of the things that we've become very interested in over the years is the connection between institutional structures and agency, and how people make use of institutions that are available to them, and whether the absence of those institutions make a difference for planning and thinking about the future. You know, the whole question of agency and structure has become quite an important and I've been in contact with other researchers recently, we've done some comparative work concerned with Germany and the United States. Helga Krüger, the University of Bremen, who was a visiting professor here a couple of years ago and I wrote a paper on the social-psychological implications of going through a situation that involves a structured transition from school to work as in the German apprenticeship system, versus the more open-ended situation that youth in the United States experienced with much less direction. These are some of the kinds of questions that we've been dealing with more recently.

Design

Q: Let's talk about research design. How did you come up with the design you did back there in the 80s, and have you refined the design since then?

A: Well, one thing that was very important to us was to have a community sample, one that was not selected for a particular problem, since we wanted to know generally what the impacts of working might be. So that was very important, and we also wanted to study to the longitudinal. And when we initiated it, we did not fully expect that it would go on for as long as it has but we wanted to at least follow youth from early in high school to some years beyond high school in order to assess both the contemporaneous implications of working while they were in school, and some of the longer-term impacts that this experience that this might have. So those two aspects, the community sample and the longitudinal design, were very important to us.

Q: Longitudinal design encompasses a monthly survey of some kind, an annual survey, and then every four years you have a larger survey or something like that?

A: We started off with annual surveys in the schools, and these were administered while the students were in classes. So this was during the school day, we brought in the surveys; this was in the St. Paul public school system. The respondents were chosen randomly from a list of those who were enrolled in the 9th grade. So each year, for the first four years we had a survey that was administered in class. After they left high school, we had also yearly mailed surveys. But the mailed surveys had a life history calendar in the front which enabled the youth to indicate for each month since we had surveyed them previously, what their living arrangements were, what their educational experiences were, their experiences in the labor force, their family formation progress and so forth. So we have a continuous record of all these different kinds of transitions that go into the movement into adulthood.

Q: So you didn't ask them to send back a monthly report?

A: No.

Q: They reported it retrospectively month by month?

A: That's right. And in the high school period we also surveyed parents because we wanted to know what they thought of the young people's experience, and whether they might have somewhat of a different perspective than the youth themselves, so that turned out to be very interesting. Also, because we were surveying the parents, we asked them about the work experiences that they had when they were in high school to see if retrospectively some other things might come out, and whether, with the benefit of hindsight, they might have a somewhat different perspective than the adolescents would have on their work. And that proved to be an interesting design feature.

Q: Have you been adding additional teenagers as you go along, or are you still studying the same panel of ninth graders that you did in 1988?

A: We've kept with the same group.

Q: And how many of these students did you lose over the course of the past 12 years?

A: Well this is a very big issue, as you know, in longitudinal studies, something that researchers worry about, because dropout from a study like this is usually not random, and that's what we found as well. It was very easy to keep them when they were in high school; by the end of the fourth year we had still 93% of the original thousand. However, as the study has gone along, we've lost some people. But we're still doing fairly well, considering what studies of this kind are able to do. By 1998 we had still 74% of the original participants. One of the difficulties we faced is that this time of life is a period when young people are moving around a lot, they're going to school, they're living in different parts of the country, they're coming home, they're traveling, they're changing jobs frequently, often changing their names and so on. So this is quite a challenge for the researcher just to keep track of them.

Q: Well, 74% seems really very high, very good for that long period of time, so you undoubtedly must have invested a tremendous amount in tracking down these students, now adults, and finding them probably all over the world. How many million dollars did you pay to do this? You don't have to be exact at all.

A: The study… I don't have the figure that I can give you exactly about how much money has gone into the tracking operation, but the whole operation, our budget is about $125,000 per year for this. But it's an extremely laborious process. We have three or four undergraduate students working on this part of this, through much of the year. We use many ways of finding the youth. One thing that we did initially is to ask the students when they were in high school about people who could help us locate them if they were to move and so we continue to do that, so we have quite an array of different people who have been designated by the young people as contacts. We also have the families, and we know often where they have located since we know the schools that they're attending, and often something about their jobs so we can locate them through various sources, but that's kind of a study in and of itself. We have a paper about tracking.

Q: Now, I know that initially you didn't plan on doing this in other communities, other states, or countries, but could you mention something about the parallel study in Iowa and then other studies that allow you to make comparisons between the St. Paul youth and other youths in other areas?

A: Well, of course, the ideal study would probably be national in scope, so that one could consider the contextual variations of the rural setting, the urban inner city, cities like St. Paul, but we didn't think we would be able to garner the resources to do that kind of work, but what we have done is as the measures were developed and I became in contact with other researchers in various parts of the country, we've tried to share our instruments as much as possible so they have been utilized in other studies and one of the more interesting comparisons that has arisen is between working in the rural and the urban context. Building on Rand Congers' work in rural Iowa, we found that employment has quite different implications for the family and for the youth's place in the family in the rural, more communalistic setting than in the urban, more individualistic setting. And we've related that to the ways the money was spent and how in the rural setting the youth's wages were much more likely to be part of the family's economic strategy, used to give money to the parents, or to buy things that the parents would otherwise have to pay for. Whereas in the urban setting, it's much more individualistic used and more leisure time activities and so on, though, even in the urban setting, a large portion of youths save their money to go to college, so it's not just all for designer jeans and rock concerts.

Measurement

Q: Before going on to more analysis, I wanted to ask about the questionnaires themselves, I believe that you've allowed the questions to change and for other colleagues to submit questions in the study; how has that worked?

A: Well, initially I wanted to utilize the measures that have been used in research on adults. Up to this point, when we embarked on this study, there had been very little research on the quality of youth work. Most people seem to think that kids' jobs were just about all the same and so there was no reason to study variation in employment because the most significant thing was the number of hours that they worked, so it was important to study that, but there was no attempt to assess the intrinsic benefits, the extrinsic rewards, the stressors in a work setting, the character of relationships with supervisors or co-workers. So what we did then was to try to draw on major studies that had been up to that time on national samples in the United States, including Melvin Kohn's work on self-direction and autonomy in the work setting, and it's implications for psychological development. That was a quality of employment survey that had been done through the University of Michigan's Institute for Survey Research. There was a study for youth and transition that Jerald Bachman had done also at Michigan, that got at some of the issues that I was interested in studying. So we drew from all of these surveys and then tried to adapt the questions to the adolescent experience insofar as possible. So that was the general strategy in developing the survey initially. Of course in the beginning, we had to do a lot of pretesting to make sure that the questions worked and the young people understood them. But as we've gone along, we've changed the survey and we've had different modules added that deal with related issues. But of course as the adolescents grow and their life circumstances change, there are new issues that become relevant and so you'd want to ask about those. For example, we added in the 1997 survey a lot of questions about their sources of income from parents, from scholarships and fellowships, from grants, from welfare, from partners, and tried to understand how young people are sort of putting together their economic livelihoods. One of the greater advantages to a study like this is that since we do have the annual data collection it presents the opportunity to build in new questions as they go along and for other investigators to contribute to the study and to make use of the data. So instead of just starting a new cross-sectional study, there's a wealth of information, so one can then examine the precursors of the different kinds of economic strategies of more or less successful economic adaptation at a given point in time, and then also, as we continue to follow the youth, to consider the consequences of these various strategies. So we've had a number of those kinds of modules. Right now there are three ongoing dissertation projects that are based on data that have been developed in the course of the last few years. So the study has really taken off and to very different directions, for example, Sabrina Oesterle, who is a graduate student here, is interested in environmental attitudes and how young people develop a sense of the environment and the importance of preserving it, or on the other hand, disregard for the environment and whether there were relationships between attitudes toward the environment and actual behaviors of recycling and other kind of environmentally-protective action. And so she developed a set of questions that she's using now for her dissertation.

Data Analysis

Q: I wanted to ask about analysis strategies, the data, since it's overtime, can be analyzed in ways that most survey data can't be. And I wondered if you would just tell us a little bit about some of the more powerful techniques you used for the analysis?

A: Well, that question could be addressed in a number of different ways. One challenge that one faces as a result of having longitudinal data, especially if your study is taking place at a time when people are changing cognitively, and developmentally, and it's not always certain that the measures that one uses at one time will still reflect the same underlying phenomena at a later time. So in our studies of various psychological traits or dispositions, including self-esteem, depressive affect, the sense of personal control, we've been very careful to assess the measurement structures of these items over time and using confirmatory factor analytic procedures to make sure that the interrelations among the items are not changing in such a way, that would suggest that what we're really dealing with a different substantive phenomenon at a later time than at an earlier time. So this is just an example of one of the kinds of issues that one faces as one does a longitudinal study like this. Another thing that is important, yet somewhat disheartening, because of the complexity involved is that so often if you are doing an analysis and you think you have a predictor and you have a dependent variable and you're assessing the relationship between them, and you do this at one point in time, and you get an answer, it's rather sobering to then try to replicate that at another point in time, and just see whether it's similar or changing. And if the points in time are very close, and one gets a different answer, then it is likely that there are measurement problems or chance variations or something that one is picking up on. So especially in the beginning of the study, I was very careful to not make conclusive pronouncements about the impacts of the various stressors and rewards and so on, or hours working, until we knew that this was a robust finding that would extend across occasions. Now another way of thinking about this is in terms of the developmental change, and at one point in time there truly may be an underlying difference in the way that variables relate to each other. But if there are not theoretical or empirical or substantive reasons to believe that that would be the case then I think the more secure kind of stance, or reasonable stance, is to then demand replication or to consider findings robust. Another way of thinking about this is that if you have longitudinal data one can often develop the kinds of measures that would be impossible if one only had cross-sectional data. And so I think this is a kind of a general issue in that if you are taking snap shots of people at any time the kind of experiences that they may be having at that time may not be characteristic of their experiences over a longer period of time. And so in the case of working, many jobs that young people have are transient; they move in and out of the labor force. What were interested in… the entire pattern over the course of high school of the jobs that they had. Now because we had snap shots of their work at every occasion that we measured them, the four occasions. But also each year we asked them about the jobs that they held since the previous time we surveyed them. We could construct on that basis a typology of work experience that considered both the duration—the number of months that they were employed during high school—and the intensity of their work; that is, how many hours they were employed on the average during the period of their employment. So this typology proved to be quite powerful in predicting both contemporaneous and subsequent outcomes. And we found that there are often was not a really good consistency between, say, how many hours they were working at the particular time of measurement and their overall pattern. So this indicates really the value of longitudinal research and some of the kinds of analytic strategies that one has available that just would not be possible with another kind of design.

Interpretation

Q: You mentioned some findings already. What other interesting findings have there been relating to the initial three main questions that you mentioned?

A: There have been a lot of findings. We have quite a number of papers that I will make available to you, and they are also available through the Life Course Center at the University of Minnesota. With respect to the three questions… in terms of the kinds of work that young people have as they go through high school. On the one hand if they were kind of suck in what one might think of as youth jobs, flipping hamburgers at a fairly low level of complexity and responsibility, then there would be perhaps more cause for concern about youth work. But if the work experience is changing over time such that it is becoming more complex, more responsible, involving more supervisory experience, and so on, then we can think of jobs as changing somewhat to match the growing developmental capacities of young people themselves and also allowing for a range of different kinds of work experiences during this very critical period in the life course. And what we did find was that comparing their work experiences over time across a variety of dimensions, that there was reason to expect, reason to think that there was a progression, in the later jobs tended to be more complex as indicated by a dictionary of occupational titles complexity ratings than earlier jobs. We asked them about the amount of training they received for their jobs, when they started training, as well as continuing training on the jobs. And the later jobs involved more of that training experience. They also had more supervisory responsibilities. So we really found a progression in that first jobs tend to be informal, for girls it is babysitting and for boys it is yard work, and shoveling snow, and that kind of thing. Then a lot of young people have restaurant types of jobs—working in the fast food industry. But then by the time they get into the eleventh and twelfth grade they span out more across a wider range of different kinds of work experiences. So with respect to the first question, we found that indeed there was variety in the kinds of work that young people do and change over time in that experience. With respect to the second question about developmental impacts, this is a question that has so many different facets that it is hard to summarize in a nutshell. What I can say in a general way, is that contrary to what most people seemed to think prior to the study, that it was the hours of work that really made a difference in terms of the impact, we found that for the most part got much more mileage out of the variables that dealt with work quality. It wasn't so much how many hours they were working or whether they were working or not, but it was the developmental kinds of positive outcomes especially concerning the crystallization of work values, and self-concept, that working had a more positive impact on the sense of internal control and mastery when there were positive work experiences of both in an intrinsic and extrinsic nature. We've interpreted some of these findings in terms of youth either moving or not moving to conceptions of what they think of as future positive possible selves. Psychologist Hazel Markus has coined the term "possible self" to refer to the kind of person you would like to be in the future, the kind of person that you imagine yourself to be. And it seems that young people, as they work during high school, are kind of getting their foot into the door, they are practicing a kind of role that is very important to their future possible selves. Now if those roles are proceeding well, if they are getting a sense that they are being appreciated in the work environment, they are being compensated well, they have opportunities for advancement, they have opportunities to discuss important decisions that have to be made with their supervisors, that there are positive developmental impacts, then if they have other kinds of circumstances that do not involve those experiences, especially if they involve contradictions and conflicts between school and work, if they feel that their jobs interfere with schooling. However, we found that in most cases, those problems were not highly prevalent. With respect to drinking, alcohol use, this was a domain where the hours mattered. The more hours they spent working the more they were using alcohol. And this is a very robust finding in this area of research. The studies at Michigan, Monitoring the Future and so on, have correlated working hours with alcohol use and drug use and smoking and so on, and find that this reoccurs in the studies over and over again. And it may also be part of this kind of more precocious moving forward into adult roles, and taking on ways of spending leisure time that are perceived by the young people as perhaps more adult like.

Q: So drinking more was not necessarily really heavy drinking?

A: No it was just more frequency. But we've found by following up the youth after they left high school that the relation of work to alcohol use is much more a matter of earlier onset rather than long-term higher use in relation to their peers. So that if we chart this over time—and I can give you some displays that depict this—the youth who worked more hours during high school are more frequent alcohol users during high school, and then they continue that after high school, than those who were less involved in the labor force were lower. But then they just catch up, so that four years after high school they are identical and there is no statistically significant difference between the two groups. So this was one area where long hours proved to be problematic. With respect to the transition to adulthood, this is an important question because if it turns out that working during high school makes you feel more like an adult, and gives them some money to spend, if it draws them away from school as the critics of youth work would contend, then this is problematic. Since it so important, as everyone knows, to receive more education in this very highly technological, and increasingly so, workplace. And so it important for young people to stay in school and to get more higher education. There have been concerns that young people who worked a lot would not only be drinking more, but they would also be wanting to start their families sooner. And early family formation and parenthood could also interfere with further higher education. So these are important questions to address. It may be considered somewhat surprising, but we found that with respect to higher educational attainment, it was not the youth who didn't work in high school that were the more advantaged. But instead it was the young people who learned to balance, and through their behavior enacted a kind of balance between school and work, that they worked during most of the months of observation during the high school period, but they limited their hours of work to twenty or less per week. And these were the young people that later had the most months of higher educational attendance. The youth who worked more hours during high school, who were consistently working more than twenty hours per week, were not as advantaged in this respect. But the young people who didn't work at all were kind of in the middle. And so it seems that working even could be considered a kind of advantage, in that if young people continue to go to school after secondary school, they are usually supporting themselves at least partially. And if they have learned to combine school and work already that it eases the transition to that new situation. So it seems that the implications of work depend a lot on the pattern in which that labor force experience takes place and also its contextual meeting. The youth who worked but then limited their hours of work were the ones who planned to go to college; they were saving their money to go to college. And so working and saving and going to school were very much part of the same pattern to achieve their higher educational goals. And we haven't found evidence that the early labor force participation is related to precocious parenting and early family formation.

Q: Your findings have a great deal of policy relevance. Yet most of your writing, I think, is towards academic audiences. How do you balance these two interests, these two audiences?

A: It is very much of a challenge to bring work like this to the attention of policy makers. But because it deals with issues that are of concern to a lot of people in educational circles, people who are concerned with labor force policy, I do find myself invited to various forums to present the research and to discuss the results. For example, there was a full year study of the implications of child labor that was sponsored by the National Academy of Sciences that I was invited to participate in. And many of the conclusions of the Youth Development Study were incorporated in their report. The report dealt with a very wide range of issues concerning child labor, including injuries and the health consequences of different kinds of labor force participation and very much concerned with issues of migrant farm-worker children, children being given responsibility to work with dangerous equipment, and issues I had not addressed in the Youth Development Study. But where we came in was considering the psychosocial impacts and whether there may be benefits to allowing youth to work. There was some sentiment on the panel that youth work should be restricted. And this did work into the recommendations for a very good reason, to illuminate the abuses that were occurring. However, I think the contribution of the Youth Development Study to that debate was to recognize that there are positive impacts of work experience, especially when it is limited in terms of its hours.

Q: You have about 75 of more reports that you have listed as coming out of the Youth Development Study. I assume that these are mostly academically oriented papers rather than papers oriented toward a popular or even a policy-making audience. Is that correct?

A: That is correct. They are mostly directed to journals. However, we also have a book that incorporates many of the findings, and I think it is written in such a way that it is accessible to a broader audience.

Q: What book is that?

A: This is the book that was published in 1996. It's a Sage book and it is concerned with adolescents, work, and family. It presents the findings that concern some of the issues that the critics have raised about the benefits or the detrimental consequences of youth work, and especially in relation to the family. But on our Youth Development Study paper and publications list, also available there are reports that have been given to various audiences, presentations to conferences of… one of them dealt with a conference of personnel workers who were interested in labor issues. So it's broader array kinds of things. And if people are interested in this study then they can find the policy-related papers on that list.

Q: You have mentioned some things that you are thinking about doing in the future. Is there anything else you want to say about your future plans or anything else you want to say about the project?

A: Well since the period of adolescence has been extended historically, some researchers now are speaking of emerging adulthood that goes on really beyond the period of time that we've traditionally thought of in terms of adolescence. Often it's considered up until the age of 19 or 20. But since young people are going to school longer, they are moving between school and work, getting work experience and going back to school, it may be quite some time before they become established in a job. And if we are interested in the long-term consequences of an experience in this earlier period on their capacity to work, to handle stressors on the job, to achieve economically and occupationally, it may be a while before we can assess what the true consequences of this experience are. So I hope that the study will continue even beyond its current funding period.

Back to script menu.