Main objectives

In recent years, both political and scientific debates have stressed the growing societal importance of adult education (European Commission 2010; Expertenkommission Forschung und Innovation 2010). This claim is mostly justified with ongoing globalization, skill-biased technological change, and the development of a knowledge society. These structural changes are having crucial effects on the working lives of the population in (post-) industrial countries: Education is no longer an asset achieved in youth that remains of constant value during a long and stable employment career without interruptions. Today and in the future, adults are having to learn continuously to keep up with flexible requirements at the workplace and to be able to find employment in different and rapidly changing fields. Lifelong learning will become more important due to demographic changes as well, because the aging of the population in Germany will lead to a severe lack of skilled employees in coming decades. One way to meet this demand is further education of adults. Thus, adult education and lifelong learning becomes an integral part of contemporary and future educational careers.

The first main objective of stage 8 of the National Educational Panel Study (NEPS) is to collect comprehensive high-quality data on adult education and lifelong learning—including data on the learning environments of these activities and the decision-making processes leading to adult learning participation. Due to the complexity of the analyzed phenomena, we have to meet several challenges: What kind of adult education do we want to include (e.g., job-related vs. private learning)? What kind of training courses are we going to consider—only courses offered by certified providers or those offered by any provider? And what kinds of fields need to be covered, for example, training of cognitive and/or noncognitive competencies or training of specific skills?

Lifelong learning is embedded in educational as well as in work-life careers. On the one hand, participation in adult education depends on, for example, specific family arrangements, time constraints, and well-being; on the other hand, initial and adult education form employment careers, family arrangements, well-being, and political participation later in life. Thus, the second main objective of NEPS stage 8 is to collect complete and detailed data on the educational, job, and family histories of adults as well as data on their subjective well-being, health, and political participation. Data on the individual life courses serves as a background variable for adult education, and—equally important—it serves as an outcome variable of educational investments at any previous point of the life course. NEPS is designed so that all respondents participating in the lower stages will eventually become respondents in stage 8, and with this stage, NEPS will provide information on the outcome of all educational efforts as well as information on continuing educational efforts during adulthood.

The third main objective of stage 8 is to collect data on domain-specific and domain-general cognitive competencies during adulthood. So far, we know little about how competencies are acquired, distributed, and changed over the life course (Allmendinger and Dietrich 2004). We aim at closing this research gap by gathering data on reading, mathematical, and computer skills as well as information on a person’s interests, self-concept, and motivation. For the first time, we will be able to analyze the development of cognitive competencies over the entire life course combined with a simultaneous evaluation of returns to formal qualifications, competencies, and employment experiences.

Figure 1 summarizes the main objectives of stage 8 within the realm of the NEPS. The overall theoretical model of NEPS has been introduced elsewhere (see Chap. 1, this volume). In Fig. 1, we simply rephrase the five pillars representing this model and superimpose our main objectives: With data from stage 8, we will be able to study participation in adult education, including effects of learning environments, prior educational activities, and migration background as well as decision-making processes that lead to adult education participation; we will be able to analyze effects of initial and adult education on various outcomes such as labor market participation and performance, well-being, and health; and we will be able to describe and elaborate on the development and the effects of competencies over the life course. Because we adopt a dynamic life-course perspective, we will be able to assess how far previous competencies, learning environments, educational decisions, a person’s migration background, and previously achieved returns reinforce educational participation, competencies, and educational outcomes over time.

Fig. 1
figure 1

Main research objectives of stage 8 of the National Educational Panel Study

Full size image

In order to achieve these main objectives and to address more specific research questions of the five pillars of NEPS, stage 8 covers the population of all adults of working age regardless of their actual employment status. It introduces a number of innovative and unique elements for the first time in a large-scale panel study by:

  • Combining economic, sociological, psychological, and educational science approaches for a truly interdisciplinary approach to adult education and returns to education in a life-course perspective;

  • Developing and adjusting measures of various theoretical constructs for adult interviews, for example, measures of learning environments, social and cultural resources, migration-specific aspects, and returns to education;

  • Introducing modularized measures of all dimensions of educational activities (formal, nonformal, and intended informal learning) as well as detailed measures of employment activities, partnerships and children, over the entire life course;

  • Applying measures of various educational outcomes, including detailed information on labor market returns, subjective well-being, health, and political participation;

  • Repeatedly assessing the individual development of cognitive competencies in a representative adult sample, including various domain-specific assessments that can be compared with competence endowments in earlier stages;

  • Introducing elements of data editing to the interview situation by applying the latest computer-assisted interview techniques in order to raise the quality and consistency of life-course data;

  • Combining a representative sample of the adult population with cohort-specific samples of earlier stages for rich and detailed life-course information on educational careers;

  • Enriching survey data with process-produced administrative employment information via social security numbers, for example, data on wages, labor market participation, and firm characteristics as well as on unemployment and active labor market policy measures.

Conceptual framework and research questions

The research questions of stage 8 are mainly informed by microsociological and microeconomic theories on education and labor market participation such as human capital theory, signaling theory, and a rational choice perspective on educational decisions. These theoretical foundations are extended by approaches from educational science and developmental psychology. The latter approaches are particularly important for understanding competence endowment and development in adult life. Until now, not only empirical work on competence development in adult life courses is scarce, but also theoretical competence models for this development stage are underdeveloped.

The life-course perspective on educational histories and adult education

The central conceptual perspective of the study is life-course research (Mayer 1990, 2009; see Chap. 2, this volume). This approach perceives the life course of an individual as a sequence of activities and events in various life domains and spheres. Life courses are understood as rule-based, dynamic characteristics of the social structure that affect numerous individuals and decide about the social positions they reach. Life courses are influenced by institutions, intentionally or unintentionally, as well as being decided by the individuals themselves—partly in a goal-oriented behavior and partly as unintentional outcomes of their actions. Life-course research theoretically and empirically aims at analyzing the dynamics of the distribution of positions and resources of individuals in a society. This perspective allows us to handle age, cohort, and period effects simultaneously and thus account for “local interdependencies” of events and conditions (Mayer and Huinink 1990). In particular, the timing and sequencing of education and training in the occupational career as well as interlinked parallel activities in other domains of life can be taken into consideration. Thus, various research questions can be derived from or linked to this framework. With a life-course perspective, we are able to analyze the way adult education is embedded in the work-related life course and in the private life course. We learn about the influence of employers and other (labor market) institutions and about the gender-specific impact of partners, children, and family arrangements in general on participation rates in adult education. We can derive research questions on decision-making processes regarding participation in educational activities, including the importance of previous educational and occupational attainment and potential path dependencies. And we are able to study cumulative returns of adult education over the life course for various labor market and non-labor-market-related outcomes.

We also refer to concepts that capture salient historical changes in life-course patterns. The most comprehensive approach, individualization theory (Beck 1986), assumes that individuals are gaining greater control over their lives in the process of modernization and accordingly pursuing a wider variety of life designs and life trajectories. This concept will be contrasted against various other, partly contradicting approaches to life-course development such as pluralization (Zapf 1991), institutionalization (Kohli 1985), deinstitutionalization (Shanahan 2000), standardization (Kohli 1985), and destandardization (Modell et al. 1976). These concepts offer powerful and theoretically informed tools for comparing the educational pathways of different cohorts. Data from stage 8 enables us to test the empirical substance of these partly contradicting theories and contrast the development of life-course patterns to overall developments in the labor market.

Some changes in life-course patterns are already apparent. The notion of “standard biographies” has lost empirical relevance (Buchmann 1989; Heinz 2003). The traditional sequence of life stages from education to work and from work to retirement is gradually being superseded by more diverse patterns in which people may reenter education after periods of work, take sabbaticals, change occupations more often during their working lives, and combine work and other activities in prolonged transitions to retirement (Jacob 2004).

The social science view on life courses is supplemented by approaches from educational science and developmental psychology. Here, a lifelong learning and competence development perspective is central for understanding educational trajectories. This perspective implies that the development of competencies is subject to stage-specific dynamics. Whereas, for example, reading literacy is a domain-specific competence during school age, it becomes a cross-curricular basic skill in vocational training, higher education, and working life. An important open question is how basic competencies develop further during adolescence and adult life and how they influence the acquisition of domain-specific competencies in these stages.

With stage 8 of the NEPS, it will be possible for the first time in Germany to answer research questions centered on participation in education, returns to education, changes in the importance of adult learning, and developments of competencies during the adult life course. To collect information on the different educational activities in which respondents have been engaged over their life courses, it is useful to distinguish these activities analytically (see, for a detailed discussion, Kleinert and Matthes 2009; Chap. 6, this volume). Education taking place in formal learning environments is institutionalized and most often leads to recognized certificates that strongly determine labor market chances in Germany. Educational activities in this respect will be collected in stage 8 by applying retrospective life-course instruments covering the whole schooling and vocational training history of respondents. In order to cover educational activities in nonformal learning environments in adult age—institutionalized shorter training courses not leading to certificates—it is important to develop and implement a clear working definition of significant adult education. Obviously, it would be insufficient to include only adult education courses that may be of relevance for some projected return in later life. Instead, we adopt a threefold strategy: We ask for all forms of training in nonformal learning environments, the exact subject area, and the initial intention to take such a course.

Beyond pure participation rates in different population segments, little is known about informal learning, that is, learning processes that are organized by the individual her- or himself. This is particularly true regarding the decisions that lead to these learning processes or their (cumulative) returns. Therefore, as far as possible, stage 8 of the NEPS will gather information on these educational activities in a standardized way. Because people have difficulties in remembering activities in nonformal or informal learning environments over a longer time span (Dürnberger 2008), the panel structure of the survey is particularly important, because it allows us to record such activities for the limited time span between two panel waves.

Survey questionnaires cannot measure unintentional learning directly. Nonetheless, we know that this form of learning is very important—not only on the job but also in the course of voluntary work and political engagement. Thus, unintentional learning and its effects in terms of competence growth will be assessed indirectly by measuring job requirements, employment experience, as well as social and political participation. Hence, we intend to include a special focus on these topics in stage 8, especially on job tasks and their development over time.

Competence endowment and development

One central aim of the NEPS project is to generate knowledge about competence development over the life course. Up to now, knowledge about competence endowment, distribution, and change in adult life is scarce in Germany. We know little about the distribution of competencies in different groups of the adult population (for a recent exception, see Wölfel et al. 2011), and we do not know how competencies change over the course of an adult’s life. With NEPS stage 8 data, we will be able to describe these distributions and developments, and we will be able to analyze factors that trigger positive developments in competencies and the acquisition of new skills. Likewise, we will be able to detect determinants of any decline and loss of skills and competencies. Domain-specific cognitive competencies such as reading and mathematics and their importance for educational success are well-researched in school, training, or higher education settings. However, we do not know whether these domain-specific competencies and other employment-related skills continue to be of relevance beyond initial education. Likewise, up to now, we have not been able to analyze how basic cognitive competencies contribute to the acquisition of domain-specific skills in working life, and how they interact in adult life.

Different competencies may also function as crucial causes for various outcomes in adult life courses. For example, competence endowment is expected to influence educational decisions in adult life, and it may contribute to the returns to education beyond educational credentials—particularly with respect to employment-related returns. Repeated comprehensive measures of competencies, educational attainment, and vocational qualifications over the life course will provide data on the correspondence and the significance of these constructs with respect to labor market returns. And the data will allow us to assess the changing relevance and interaction of competencies, credentials, and qualifications throughout an adult’s work life.

As a first step prior to the actual measurement of competencies in stage 8, we have to select applicable domains that we want to follow during adulthood. Two considerations are helpful for arriving at a reasonable selection: First, they should be relevant for the major part of the adult population and labor force. Second, their assessment should be valid and reliable (Kleinert 2005). The first prerequisite for selection is especially risky with regard to the highly heterogeneous target group covered in stage 8. Furthermore, the selected competencies should be compatible with the competencies measured in the earlier educational stages. For this reason, it will not be possible to answer the question of competence selection in stage 8 independently of the other stages, and this work has to be based on an overarching competence design.

As illustrated in Chap  5 in this volume, one focus of stage 8 (as well as of the earlier educational stages) will be on cognitive competencies that are domain-specific during schooling but basic in adult life. It is undisputed that competencies such as reading and mathematical literacy, scientific literacy, and foreign-language literacy are necessary prerequisites for successful employment participation and active participation in society (Rychen and Salganik 2003). Taking our heterogeneous target group into account, it is important to adapt the existing student assessments to cover functional literacy, that is, to assess possible problems in adult daily life, and to build tests measuring the full range of competence domains in the adult population. Another important demand is to cover the dynamics of these competence domains over the adult life course.

A second important focus in stage 8 is on measuring skills connected to the employability of adult individuals, that is, competencies that help adults to find and maintain employment in different occupational fields under changing conditions. In this sense, different key competencies or metadisciplinary skills are supposed to play an increasingly important role in adults’ lives. With regard to this concept, we intend to measure selected noncognitive competencies indirectly by self-assessment scales (see Chap. 10, this volume). Because educational processes in adult life are mostly self-directed, we consider subject-specific interests, self-efficacy, self-regulation, and motivation to be particularly important, and we will include these concepts in the instrument—as well as more stable personality traits such as the Big Five. Furthermore, social behavior and cooperation are usually considered important in adult life, especially in working contexts. Due to the large amount of preliminary research and development, certain facets of social competencies will be included in later panel waves of stage 8. On the cognitive side, literacy in information and communication technologies (ICT) seems to be particularly important for many tasks in employment contexts as well as in private life. On the one hand, ICT literacy has unique cognitive and technical aspects; on the other hand, it serves as a “tool” for applying other cognitive and social competencies (e.g., writing texts or communicating with others). Most important in our context is the fact that ICT skills are relevant for employment chances beyond specialized occupational fields. Finally, measures of metacognition will be included, i.e. knowledge, skills, and attitudes that help to make strategic decisions when learning or thinking and to initiate, organize, and control their active realization. Metacognition should be highly relevant especially for lifelong learning processes.

Learning environments

Learning environments differ substantially for respondents in stage 8, especially when compared to students and children in earlier educational stages. Whereas learning in all lower stages of NEPS takes place in the same predefined institutional contexts for the survey respondents (but see Chap. 12, this volume), the learning processes of adults occur in a multitude of different learning environments.

Concerning learning in formal and nonformal environments, Germany has numerous providers offering training or courses of adult learning, for example, firms, state-founded institutions such as adult education centers, state agencies such as the Federal Employment Agency, chambers of commerce and crafts, higher education institutions, and a wide range of non-governmental organizations (Kleinert and Matthes 2009). It is difficult to gain a complete overview of the relevant providers of adult learning, particularly since the country’s federal structure complicates the field substantially. Programs vary considerably across the federal states, and it is impossible to identify a coherent top-down approach to adult learning policy. Thus, we need to collect individual basic characteristics of the training programs and of the providers of adult education, that is, standardized information from the adult learner’s perspective in order to answer questions on the effects of the structural characteristics of learning environments on different educational outcomes.

Most adult learning does not take place in standard educational institutions (like schools or universities). In this respect, learning environments in stage 8 differ considerably from other educational stages. The most important environmental context of adult learning is the firm or workplace. Employers provide a substantial part of further education and training for employed adults in Germany (Rosenbladt and Bilger 2008). In general, large firms provide considerably more training than small and medium-sized firms. They provide a great deal of training themselves, whereas small and medium-sized enterprises often turn to external providers. Furthermore, participation in formal adult education relates strongly to employment patterns. Individuals who are employed in small firms, in temporary jobs, or part-time are less likely to enroll in formal adult learning. In this respect, it is worthwhile to connect research on training patterns to theories of labor market segmentation (cf. Sesselmeier and Blauermel 1997).

Another perspective of adult learning in firms is to gather information on “learning by doing” at the workplace by investigating job requirements. The idea behind this approach is to ask in detail about the tasks required by the job in order to gather information on skill demands, learning possibilities, and learning conditions in different workplaces. Thus, the supply-centered main perspective of education in NEPS is complemented with information about skill demands. It allows us to answer a broad range of questions, for example: How do job requirements, formal education, and competencies match in the adult population? Which factors decide about under- and over-qualification? How do job requirements change in changing labor markets, and what skills will be required in the future in order to guarantee stable employment careers?

Besides firms, the individual household serves as an important learning environment for adults. Household involvement in learning processes can take a number of different forms, especially the creation of a positive learning environment at home. Other household members, in particular the partner, may provide economic, cultural, social, or time resources for the investment in education or they may hinder participation by denying them. An analysis of family and household effects on adult learning participation may be particularly promising from a gender perspective. Up to now, we know that women, in particular mothers, participate less in further training than men or women without children, but the dynamics leading to these differences have not yet been fully understood (Dieckhoff and Steiber In Press).

Alongside structural information on adult education providers, firms, and households, an important feature of stage 8 is to gather information on the specific characteristics of adult learning courses. Based on a general model of how a course is conducted, of the atmosphere of a course, and of the cognitive activation it triggers (see Chap. 6, this volume), we collect data on these dimensions for selected courses. We cover the setting of a course as well as its internal design as indicators for the structure of a course. The support dimension captures both interaction patterns between participants and instructor and interaction patterns among course participants themselves. Last, we ask participants of adult learning courses about the cognitive challenge they face when taking the course. With detailed knowledge on these characteristics of adult education courses, we will be able to analyze the effects of the characteristics on successful participation, on increases in skills or competencies, as well as on returns to adult educational investments.

Social inequality and educational decisions over the life course

Unlike students in lower stages, respondents in stage 8 are not institutionally forced to make educational decisions at particular points of time in their life courses. The main research question for adults, therefore, is why adults engage in (or refrain from) education, and if they engage, in what types of adult education they engage. Educational decisions might be based on a rational decision process, on heuristics taking only limited information into account, or a mixture of the two. We collect data that will allow us to test various theoretical approaches regarding educational decisions, the most prominent being rational choice theory (Erikson and Jonsson 1996), satisficing (Simon 1993), models of frame selection (Esser 2001), simple heuristics (Gigerenzer and Todd 1999), as well as other approaches promoting the idea of bounded rationality (e.g., Brewer 1988; Chaiken and Trope 1999; for more details on these approaches, see Chaps. 7 and 10, this volume).

In order to understand educational decisions, it is crucial to have data on adults’ work-related and non-work-related aspirations. In case of work-related aspirations, it is essential to collect information on attitudes toward labor force participation, working hours, workload, family, division of domestic labor, and occupational career aspirations. Likewise, in order to understand adult education investments, it is essential to gather information on attitudes not related to the realm of the labor market such as attitudes and aspirations regarding leisure-time activities and measures of self-concept in general. Predominantly, participation in adult education is understood as a means to meet aspirations or to live up to specific attitudes. Thus, theoretical models explaining participation in adult education have to focus mainly on attitudes, benefits, and probabilities related to the prospected returns to adult investment, that is, returns related to the labor market (e.g., career outcomes) or to private interests.

Interpersonal influence processes are of high relevance for participation in adult learning. This holds all the more because educational decisions and the reasons for them vary between different social groups (Boudon 1974; Breen and Goldthorpe 1997). Members of different social groups might have different attitudes and information, and their educational behavior may be affected by different economic, cultural, and social resources. A lack of financial resources to pay the fees for a course or to make up for opportunity costs has an impact on the odds of engaging in further education. Cultural resources, especially previous educational achievements, as well as competencies influence educational decisions significantly. Social resources are crucial in order to learn about adult education offers, to gain active support from other course members, to receive active support in the household in order to have time to take the course, to acquire financial support, and, of course, for educational aspirations in general. Social background and the interplay of primary and secondary effects are particularly interesting for the case of educational decisions in adult life because of the competing explanations for the declining effect of social origin (Hillmert and Jacob 2005).

In general, participation in adult education depends on the initiative of individuals. However, employers also play an important role with respect to educational decisions, because they exert a strong influence on the decision-making processes and participation chances of their employees. It is necessary to study who is encouraged and promoted by an employer to take part in further education and who has no access to further training on the job. Besides the firm, the family situation (as part of the social resources) shapes preferences and chances to engage in adult education. In this respect, educational decision-making differs greatly from educational situations at student age, due to the lack of institutionalized decision-making processes and the individual context of firms and family situations. Thus, existing theoretical approaches have to be adapted or supplemented in order to adequately explain educational decisions in adult life. This applies in particular to the role of significant others, as social resources are not provided by parents, but by partners, friends, and colleagues.

Optimal and actual timing in the context of the life course is another important aspect of adults’ educational decisions. Information on time preferences gathered in the NEPS will allow analyses of such decisions beyond mere cost-benefit analyses. Unlike the earlier education stages covered in the NEPS, the respondents in this stage will not necessarily be participating in or preparing for imminent educational activities at the time they are being surveyed. Nevertheless, several indicators allow us to analyze the processes that lead to educational decisions. These indicators can be measured for all participants independently of whether they plan to enter education or training. Thus, it is important to gather information on educational aims, attitudes, and expectations; information on the motive of avoiding downward social mobility; and information on and knowledge of educational opportunities. Likewise, we collect data on perceived benefits of educational investments, perceived probabilities for a successful investment, and on projected costs of participation in adult learning. In addition, it is important to ask respondents about their overall financial, cultural, and social resources. In further panel waves, these determinants can be evaluated against possible educational decisions and their outcomes.

Special target groups: migrants

Detailed knowledge of further education in the adult population with migration background is scarce in Germany. At the same time, migrants and their descendants are a group that is at least partly in more need of adult education because their educational background is often inadequate and/or their certificates are not transferable to Germany. Thus, an important objective of the study is to generate empirical findings on migrants’ competencies and literacy in the German language; their financial, educational, and ethnic resources; their participation in further education; and the returns to their educational and occupational investments.

For this purpose, the survey will gather detailed information about migration background up to the third generation. Respondents with migration background will be assessed in German language literacy, like all stage 8 respondents. In addition, we ask for their native languages as well, for language use in the household the respondents grew up in and in the current household, and for language use at the workplace and in leisure time. Furthermore, we survey the complete migration history of the respondents, including what kind of legal status they had when they came to Germany. We are also interested in migration-specific cultural and social capital and the effects of ethnically homogeneous or heterogeneous networks on the odds to participate in adult education and the chances to succeed in different labor market segments. For a valid measure of human capital, we are not just asking respondents with migration background to translate their educational degrees into an equivalent degree in Germany. The two largest groups of migrants (migrants with Turkish and Former Soviet background) are invited to write down the actual name of their original degree in Turkish/Russian, and we ask whether their degree has been recognized by German authorities. Furthermore, we offer these two migration groups to interview them entirely in their native tongue with a translated questionnaire (for more details, see Chap. 8, this volume).

Returns to education

The last main objective of stage 8 is to collect information on returns to education. Economists are mainly interested in the effects of education on wages or unemployment risks (e.g., Lauer and Steiner 2001), whereas sociologists more often refer to class outcomes, job prestige, or occupational mobility (e.g., Shavit and Müller 1998; Scherer 2005). The same is true for the effects of continuing training. Economists focus on wages or unemployment risks (Jenkins et al. 2003; Kuckulenz 2007), whereas sociologists also address the influence on vertical or horizontal occupational mobility (Dieckhoff 2007; Wolter and Schiener 2009). These returns can be observed only when persons have left initial education and belong to the active or passive labor force population. Thus, it is essential to extend the NEPS beyond student cohorts to the adult population in order to address research questions on returns to education.

For a number of reasons, economic and occupational returns to education can be better analyzed with data from stage 8 than from most other existing adult surveys, for example the German Socio-economic Panel (GSOEP). First, we are able to investigate the returns to different educational activities in different life phases, enabling clear temporal modeling. Second, we provide detailed measures of educational degrees, fields of study, and adult education in formal, nonformal and informal learning environments, surpassing the measures of all existing data on education so far. Third, we are able to differentiate between returns to educational credentials and returns to competencies. Human capital theory (Becker 1964; Mincer 1974), signaling theory (Spence 1973), screening theory (Arrow 1973), and the job competition model (Thurow 1975) make different statements on the relevance of certificates and competencies for labor market success. However, until now, it has hardly been possible to compare these theories empirically because applications are usually based on the same limited qualification indicators and competence measures are usually lacking (cf., for exceptions, Green and Riddell 2003; Tyler 2004). Fourth, with the explicit permission of the respondents, we add administrative social security information to their files, including highly reliable measures of wages and labor force participation for their entire working life. Finally, our data contains detailed information on social origins, migration background, age, and gender that allows us to analyze returns to educational degrees and competencies specifically for various well-defined subgroups of the population.

Educational achievement also contributes to explaining disparities in several life spheres apart from working life and the economic domain. For this reason, our concepts of returns to education go beyond questions of pure economic applicability. We are able to investigate how education shapes competencies later in life, and how education and competencies affect social and political engagement, subjective well-being, and health. Subjective well-being is selected because it contributes to a more holistic concept of the social welfare returns to education than the purely economic aspects of wealth (see Chap. 9, this volume). By analyzing social and political participation, it is possible to draw inferences on not only private but also social returns to education. With such external effects, we are able to derive more comprehensive utility functions underlying the decision whether to invest in human capital or not. Additionally, these dimensions are selected because they both fulfill additional functions in the context of stage 8. Whereas well-being is an important determinant of motivations and aspirations, knowledge on social participation helps to explain individual competence endowment, because competencies (in particular, social and personal) may be acquired in the context of voluntary activities (cf. Gerzer-Sass et al. 2006; Kirchhöfer 2000).

Methodological aspects

Survey design and survey modes

The main aims of stage 8 are to provide data on (a) adult education and lifelong learning, (b) economic and noneconomic returns to education, and (c) the development of competencies over the life course. Because adult education is often not institutionalized, and learning can take place in almost any circumstances and life phases, stage 8 requires an individual sampling strategy and a broad target population. We define our population as all adults of working age living in Germany, irrespective of their labor force participation. To separate our target group from the earlier stages in NEPS, we consider only persons who are usually not enrolled in initial training anymore—at least if they have low educational attainment. Thus, our starting population for stage 8 comprises adults aged 23–64 years. A description of the sampling methods of stage 8 is provided in Chap. 4 in this volume. The panel design of stage 8 stands out within the NEPS framework because the first panel wave already started in 2009. Starting from there, we follow up participants at yearly intervals.

In order to test domain-specific and domain-general competencies in the adult population, we are going to test our respondents in one or two competence domains every second year. At present, these tests are self-administered paper-and-pencil tests. In these survey waves, we will also update the life course of the respondents for the time since the last interview and ask some additional short questions from other NEPS pillars. An interviewer will visit the respondents at their homes, conduct the assessment, and carry out a computer-assisted personal interview (CAPI). In odd-numbered waves, the main focus is to gather information on constructs provided by pillars 2–5 (including concepts of interests, self-concept, and motivation) and stage 8 as well as to update the life course. This information is usually gathered via computer-assisted telephone interviews (CATI) due to cost restrictions.

In detail, our concept of survey modes is more flexible, but also more complex. All sample members without telephone number information are visited by interviewers at their homes to gather their telephone numbers. If a respondent refuses to participate on the phone, a CAPI interview is conducted. Likewise, if a respondent refuses to welcome an interviewer at home, we offer a CATI interview and waive the competence assessment. Thus, we offer a deliberate CATI/CAPI mix in order to minimize nonresponse bias and panel attrition. The interviews last about 30 minutes when combined with competence assessments and at most 60 minutes in waves without assessments. The length of the competence tests varies between 30 and 60 minutes.

Questionnaire design

Regardless of the survey mode, the interviews in stage 8 are always computer-assisted. Computer-assisted interviews provide numerous opportunities to enhance data quality and to conduct customized interviews due to the nearly unlimited filtering possibilities. We make use of these opportunities to a large degree. Most important are technical innovations in the life-course instruments, because a significant part of the yearly interviews is devoted to collecting or updating life-course information. In order to receive reliable and valid information in this field, all subareas of the life course are recorded as independent longitudinal modules. These modules are general schooling, vocational preparation schemes, vocational training and higher education, military service, employment history, unemployment history, partners in the household, children, and parental leave. Within these modules, individual episodes are recorded in chronological order. This modular design guarantees adequate reporting of the complexity of present life courses.

Two technical strategies are combined to assist the respondents in remembering and anchoring their activities throughout time: First, interviewers can access information respondents have already given earlier in the interview to help them date other events correctly, for example, by recollecting the date of graduation when asking for a date of relocation. Second, after initial data collection, the time consistency of dating in all modules will be checked for coherence in a special data revision module. This is done together with the respondents, so that mistakes and problems can be corrected immediately. This step integrates immediate data editing into the inquiry, so that any inconsistencies that emerge can be clarified together with the respondents during the interview. Successful implementations of this design in the most recent wave of the German Life History Study (Hillmert et al. 2004) and in the forerunner study of stage 8, Working and Learning in a Changing World (ALWA) (Kleinert et al. 2008; Antoni et al. 2010), have shown that the quality of data rises and significantly more episodes of education and unemployment are reported (Drasch and Matthes In Press).

Conclusion

This short overview has hopefully ignited the reader’s interest in the theoretical approaches and in the data generated in stage 8. The three main objectives of stage 8 are to provide rich data from the life-course perspective on adult education, on competencies, and on labor market and non-labor-market-related returns to educational investments. We are eager to collect data that enables testing different theoretical models, and we therefore invite researchers of different disciplines and different research paradigms to come up with bold research questions that may be analyzed with stage 8 data. Given our detailed data collection on retrospective life-courses and process-produced employment information, many research questions can be addressed as soon as the data from the first survey wave is released. For some questions, researchers may have to wait for some years until enough data has been collected in future panel waves. And finally, the design of stage 8 allows for the assessment of long-term developments once the younger cohorts have been transferred into the survey program of stage 8. Thus, we are setting up a data collection program that will provide detailed high-quality information on numerous aspects of the (educational) life course for many years to come. With our program, we hope to stimulate a stream of fascinating research questions and findings for the adult population in Germany.