Social Sciences
Longitudinal Studies
Longitudinal studies are research designs that involve collecting data from the same subjects over a period of time, allowing researchers to observe changes and development. These studies are valuable for understanding how individuals or groups change over time and for identifying patterns and trends that may not be apparent in shorter-term studies.
Written by Perlego with AI-assistance
Related key terms
1 of 5
10 Key excerpts on "Longitudinal Studies"
eBook - ePubResearch Design
Succesful Designs for Social Economics Research
- Catherine Hakim(Author)
- 2012(Publication Date)
- Routledge(Publisher)
8 Longitudinal Studies DOI: 10.4324/9780203354971-8 Regular surveys and Longitudinal Studies are the two designs that most explicitly focus on social change processes, but they do so in different and complementary ways. Regular surveys provide information on net change at the macro-level, while Longitudinal Studies provide information on the much larger volume of gross changes (or flows) at the microlevel (see Chapter 7). Longitudinal Studies are often initiated when regular surveys or other sources reveal new trends that they cannot fully describe or explain (especially if they do not contain a panel element). The longitudinal study is unique in its ability to answer questions about causes and consequences and hence to provide a basis for substantiated explanatory theory. The prospective longitudinal study takes a single sample or group and follows it up, with repeated data collections, over a long period of time (see Figure 1). Strictly speaking, the term longitudinal study refers only to prospective studies. The retrospective study, which is quicker and cheaper, is considered below as a quasi-longitudinal design. Longitudinal Studies cover a ‘long’ period of time, although what is considered a long period of observation depends on the subject-matter and context, and the issues addressed. Studies of people experiencing a spell of unemployment may continue for only one or two years, while studies of human growth and maturation may continue for 30 years or longer. Generally, there are numerous data collections in order to collect information on changes as they are happening
eBook - PDFDevelopment in the Preschool Years
Birth to Age Five
- Thomas E. Jordan, Allen J. Edwards(Authors)
- 2013(Publication Date)
- Academic Press(Publisher)
Longitudinal Studies 2 This chapter will present information about Longitudinal Studies, a do- main of research that has become popular once more. The resurgence of interest is recognition that longitudinal data provide a way to understand important aspects of human situations. For students of development the first important data set is the set of growth measures gathered on his son by M . Gueneau de Montbeillard. The measures began with the boy's birth on 11 April 1 7 5 9 , and ended on 11 November 1 7 7 6 , when the boy was 16 years and 7 months (16:7). The values are recorded in the pre-metric units pieds, pouces, and lignes (foot, thumb [inch], and cord). This interesting data set was not published by de Montbeillard, but is extant because it appears in Sonnini's (1799) edition of the works of his colleague, Georges le Clerc, Comte de Buffon. De Montbeillard had previously helped Berryat with his 1 7 5 4 edition of the early proceedings of the Academie Royale des Sciences. Today, longitudinal study exercises the minds of people in a variety o f fields. In addition to studies in child development, there are major inquiries in schizophrenia (Sartorius, Jablonsky, 6c Shapiro, 1 9 7 7 ; Mednick, Schul- singer, & Venables, 1 9 7 9 ) , occupational choice (Parnes, 1 9 7 5 ) , and educa- tion (Fetters, 1 9 7 5 ) . In the abstract this is not surprising, since data over time can lead to fascinating insights. Concretely, there are many practical problems, and the challenge to organizational skill can be enormous. In the past, people used samples that were selected in part because they seemed 7 8 2. Longitudinal Studies easy, comparatively speaking, to study. For example, the Oakland-Berkeley studies used protestant, middle-class white families. With equal facility, investigators have chosen to drop certain kinds of children as objects of study (probands).
eBook - PDF- Brett Laursen, Todd D. Little, Noel A. Card, Brett Laursen, Todd D. Little, Noel A. Card(Authors)
- 2012(Publication Date)
- The Guilford Press(Publisher)
There are differences between Longitudinal Studies in their overall productivity and publishing outlets. Some Longitudinal Studies have been reported mainly in books, some in journal articles. In the case of several collaborators, it has to be decided who is involved in publication and how to avoid publishing overlaps. A related issue is the question of how long to wait before publishing the results of a longitudi- nal study. Data collection for the whole duration of the study may take years, so it may be necessary to publish results based on separate data collection phases before all phases are completed. In addition, researchers have to consider how to effectively communicate the findings to a general public, the media, and those who may apply results in practice. Summary The longitudinal design in the study of human development has established its position in science gradually during the past hundred years. The concept of development has affected the investigation of human development longitudinally, but Longitudinal Studies have also affected the concept of development. Earlier studies were focused on children and ado- lescents, but when the participants became older the studies were extended into adult- hood. Consequently, Longitudinal Studies have increased understanding of developmental processes, not only in childhood and adolescence but also in adulthood and old age. Studying individual development longitudinally is an attractive approach to research, but conducting a longitudinal study sets demands that should be critically considered in the planning stage of the study. Important considerations include, for example, a long-term commitment to the study, funding, sampling, estimating the duration and needed frequen- cies of data collection and the feasibility of tracing the sample over time, the availability of relevant measures across the studied ages, and the sustainable quality of the data. The first
eBook - PDFChildhood Poverty
Multidisciplinary Approaches
- Oxford Department of International Development, Michael Bourdillon, Jo Boyden, Oxford Department of International Development, Michael Bourdillon, Jo Boyden(Authors)
- 2011(Publication Date)
- Palgrave Macmillan(Publisher)
Part I Methodology 15 1 Doing Longitudinal Research: Opportunities and Challenges in a Study of Childhood Karen Brock and Caroline Knowles Young Lives has been introduced as a longitudinal study that follows two cohorts of children in poor communities in Ethiopia, India, Peru, and Vietnam as they grow into young adults. In this chapter, after pointing to the importance of Longitudinal Studies in the social sci- ences, we use the experience of designing and implementing Young Lives to reflect on some issues surrounding the process of such studies, and on the use of their findings to inform policy. The power, potential, and challenges of longitudinal research The power of longitudinal research lies in its capacity to illuminate patterns of change in the lives of selected groups of people. Making repeated, structured observations about the same group over time allow the exclusion of unobservable individual characteristics that do not change over time (for example, an adventurous personality with a ten- dency to take risks) and the identification of short- and long-term pat- terns of change. A classic example of this can be found in the British Doctors Study, a longitudinal study which surveyed 40,000 British doc- tors six times between 1957 and 2001, and found the first statistical proof that tobacco smoking increases the risk of lung cancer (Doll et al. 2004). In the social sciences, longitudinal research can be divided into repeated cross-sectional and cohort studies. The former sample a cross- section of the population and survey it at given points in time. The latter track a group of people (a ‘panel’) selected because they have 16 Karen Brock and Caroline Knowles experienced the same event – typically birth – during a specified time period. Studies of these types can provide a glimpse into both the life histories of the individuals who make up a segment of the population, and the broader patterns of change that make up the social landscape.
eBook - PDF- Roger Wimmer, Joseph Dominick(Authors)
- 2013(Publication Date)
- Cengage Learning EMEA(Publisher)
CHAPTER 8 LONGITUDINAL RESEARCH CHAPTER OUTLINE Development Types of Longitudinal Studies Panel Studies Special Panel Designs Analyzing Causation in Panel Data Combining Qualitative and Quantitative Data in Longitudinal Research Longitudinal Research on the Internet Longitudinal Design in Experiments Summary Key Terms Using the Internet Questions and Problems for Further Investigation References and Suggested Readings 224 Copyright 2012 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it. Most of the research discussed to this point has been cross-sectional. In cross-sectional research, data are collected from a represen-tative sample at only one point in time. Lon-gitudinal research, in contrast, involves the collection of data at different points in time. Although longitudinal investigations are rel-atively rare in mass media research, several Longitudinal Studies have been among the most influential and provocative in the field. Of the 14 studies Lowery and DeFleur (1995) consider to be milestones in the evo-lution of mass media research, four involve the longitudinal approach: Lazarsfeld, Berelson, and Gaudet ’ s The People ’ s Choice (1944), which introduced the two-step flow model; Katz and Lazarsfeld ’ s Personal Influ-ence (1955), which examined the role of opinion leaders; the Surgeon General ’ s Report on Television and Social Behavior , particularly as used in the study by Lefkowitz, Eron, Walder, and Huesmann (1972), which found evidence that viewing violence on television caused subsequent aggressive behavior; and the 10-year update of the Lefkowitz et al.
eBook - PDF- Michael Brambring, Friedrich Lösel, Helmut Skowronek(Authors)
- 2015(Publication Date)
- De Gruyter(Publisher)
However, there are solutions to both problems. More specifically, it is of crucial importance to design Longitudinal Studies in a way that: (1) there remains sufficient time for analyzing the data and writing up reports between two adjacent measurement points, and (2) the design is flexible enough to allow for the possibility of change. Accordingly, an extreme delay in publication can be avoided if sufficient time is available for the different phases of a longitu-dinal study (i.e., data collection, data analysis, report writing). With regard to flexibility, precautions should be taken to ensure that the study is not too narrow in scope. That is, a broad-band investigation including several measures from different domains copes with the problem of timeliness in that the data Problems of Longitudinal Studies 317 might be reanalyzed at a later date and interpreted in the light of theoretical and technological advances (cf. Block and Block, 1980, 1984; Harway et al., 1984). All in all, there is no doubt that there are several practical difficulties with Longitudinal Studies. However, as there also seem to be practicable solutions to most of the problems discussed in this section, those problems should not be overestimated by researchers interested in conducting longitudinal research. 2.2 Conceptual Problems All five rationales of longitudinal research listed above referred to the assess-ment of change. At first glance, there seems to be no problem with conceptual-izing and studying developmental change. A closer look at the literature, how-ever, reveals that conceptualizing human development is a complicated issue. As emphasized by Baumrind (1987), instability and discontinuity in human development can only be seen against a background of stability. The question of what stability can mean in the context of changing individuals (Wohlwill, 1980) has been answered differently by different researchers.
eBook - PDFHandbook of Longitudinal Research
Design, Measurement, and Analysis
- Scott Menard(Author)
- 2007(Publication Date)
- Academic Press(Publisher)
Part I Longitudinal Research Design This page intentionally left blank Chapter 1 Introduction: Longitudinal research design and analysis Scott Menard 1 Longitudinal and cross-sectional designs for research As described in Menard (2002), longitudinal research designs can best be understood by contrasting them with cross-sectional research designs. In a purely cross-sectional design, data are collected on one or more variables for a single time period. In longitudinal research, data are collected on one or more variables for two or more time periods, thus allowing at least measurement of change and possibly explanation of change. There are some designs which do not fall neatly under the definition of pure cross-sectional research or longitudi-nal research. One example is research in which data are collected for different times for dif-ferent cases, but only once for each variable, and the time dimension is ignored. This design may be used, for example, when data are not all available at the same time, as in Ahluwalia’s (1974; 1976) study of economic development and income inequality. Although the data come from more than one time period, the design for any given case, and also the analysis, is cross-sectional in nature. The danger here lies in assuming that relationships are constant over time; the alternative is that any bivariate rela-tionship may reflect not the relationship one would obtain if all of the data were measured for a single period, but may instead be contami-nated by changes in that relationship over time. Another possibility is a time-ordered cross-sectional design, in which each variable is mea-sured only once, but variables are, by design, measured at different times. An example of this is the study by Tolnay and Christenson (1984), who deliberately selected variables which were measured at different times for use in a causal path analysis of fertility, family planning, and development.
eBook - PDF- Peter Lynn(Author)
- 2009(Publication Date)
- Wiley(Publisher)
Some of these surveys are carried out only once every n years, where n may take a value like 5 or 10. But some do indeed take place every year and thus, over time, they in principle cover the entire population of persons who have experienced the particular school system or educational establishment that is being sampled. 1.3 STRENGTHS OF LONGITUDINAL SURVEYS Longitudinal surveys present a number of options, both for data collection and for analysis, that are either simply impossible with cross-sectional surveys or cannot be achieved in a satisfactorily accurate or reliable way. Often these features are key elements of the rationale for carrying out a longitudinal survey. However, it is important to distinguish between longitudinal surveys and longitudinal data . Longitudinal surveys are a source of longitudinal data, as the resultant data include items that refer to different points in time. But there are also other ways of obtaining longitudinal data, including diary methods and the use of retrospective recall within a single survey instrument. We briefly discuss here the strengths of longitudinal data, but our particular focus is on the strengths of longitudinal surveys. 1.3.1 Analysis Advantages It is of course artificial to separate analysis advantages of longitudinal surveys from data collection advantages, as the reason for collecting a certain type of data is in order to be able to carry out certain types of analyses. The key advantages of longitudinal data 4 METHODS FOR LONGITUDINAL SURVEYS (which in most cases can only be accurately or adequately collected by longitudinal surveys) are analytical: Analysis of gross change . Analysis of gross change is perhaps one of the most common objectives of longitudinal surveys. Repeated cross-sectional surveys can be used to estimate net change, for example change in the proportion of employees who reg-ularly cycle to work.
eBook - PDFAnalyzing Social and Political Change
A Casebook of Methods
- Angela Dale, Richard B Davies, Angela Dale, Richard B Davies(Authors)
- 1994(Publication Date)
- SAGE Publications Ltd(Publisher)
2 From Cross-Sectional to Longitudinal Analysis Richard B. Davies Paradoxically, cross-sectional analysis provides the main focus of this chapter. This stems from our confident belief that most social scientists courageous enough to confront the complexities of longitudinal data analy- sis will have already accumulated some experience of cross-sectional analy- sis. Cross-sectional analysis is consequently a sensible starting point for developing the motivation required to tackle the generally unfamiliar and admittedly often difficult statistical problems presented by longitudinal data. In this context, a glib claim that longitudinal data analysis is important because it permits insights into the processes of change is inadequate and certainly fails to convince many social science researchers who are concerned with substantive rather than methodological challenges. Cross- sectional analyses, they may argue, are not necessarily uninformative about the dynamics of social change because historical information often collected in cross-sectional surveys is readily incorporated. Questions about place of residence one or more years earlier are common in popu- lation censuses and permit comparison between the characteristics of migrants and non-migrants; data on length of unemployment are usually obtained in cross-sectional surveys of the unemployed and may be used as an explanatory variable in studying attitudes, health, or motivation; and even just birth dates of children provide a demographic record which may be used in a cross-sectional study of the employment status of married women. To develop the necessary motivation, it is therefore appropriate to consider in more detail the limitations of cross-sectional analysis, and this is the first objective of this chapter.
eBook - PDF- Todd D. Little(Author)
- 2023(Publication Date)
- The Guilford Press(Publisher)
In this section, I discuss and highlight some alternative ways to consider time that will assist you in thinking about the design of a longitudinal study. If done 2 Design Issues in Longitudinal Studies with a bit of forethought, a properly designed longitudinal study will improve your ability to capture the change process that you desire to model. For a very rich but challenging discussion of the issues that I touch on here, you can try to find a copy of Wohlwill’s (1973) classic book, The Study of Behavioral Development. More recent (and more approachable) discussions of the importance of mapping theory to meth- ods and methods to theory have emerged (see, e.g., Collins, 2006; Jaccard & Jacoby, 2020; Lerner, Schwartz, & Phelps, 2009; Ram & Grimm, 2007). Most developmentalists have been schooled that “behavior is a function of age:” B = ƒ(age). This way of thinking about developmental change processes is somewhat limited, however. First, age is not a causal variable and is really only a proxy of the multitude of effects that covary with maturation and experience (Wohlwill, 1973) and are partner effects in the context of influences like race/ethnicity, sexual orien- tation, and gender. In fact, age is probably best considered as an index of context. The context of behavior in the life of a 12-year-old Hispanic gay boy would have dif- ferent connotations than the context of behavior in the life of a 16-year-old African American transgendered girl. In addition, the historical timing of measurements is often neglected as a contextual impact. The widespread impact of social media and constant connectivity has even emerged as a new field of study, what Nilam Ram has described as Screenomics. Second, as Wohlwill notes with regard to age functions, “the particular form of this functional relationship is rarely taken seriously, let alone given explicit expres- sion” (p. 49).
Index pages curate the most relevant extracts from our library of academic textbooks. They’ve been created using an in-house natural language model (NLM), each adding context and meaning to key research topics.
