CROSS-SECTIONAL RESEARCH

Longitudinal Research, Panel Study

Cross-sectional research is research which makes observations at only one period in time. Examples include conducting a survey or opinion poll. Cross-sectional research is analogous to taking one still picture of the population or group being investigated. Longitudinal research, on the other hand, makes more than one set of observations and can be compared to a simple moving picture.

In cross-sectional research, data are collected from the research participants at a single point in time or during a single, relatively brief time period (i.e., a period long enough to collect data from all of the participants selected to be in the study). The data are typically collected from multiple groups or types of people in cross-sectional research. For example, data in a cross-sectional study might be collected from males and females, from people in different socioeconomic classes, from multiple age groups, and from people with different abilities and accomplishments.

The major advantage of cross-sectional research is that data can be collected on many different kinds of people in a relatively short period of time. Cross-sectional research has several weaknesses. One disadvantage is that it is difficult to establish time order (condition 2 of the necessary conditions for causality). If you collect data from research participants at a single time point only, you can’t directly measure changes that are occurring in them over time.

Time order can be partially established in cross-sectional research through theory, through past research findings, and through an understanding of the independent variable (e.g., you can safely assume that an adult’s biological sex occurs before the amount of education they have because biological sex is set at birth). These techniques for establishing time order are weaker than actually observing people over time. A related disadvantage is that the study of developmental trends (changes in people as they get older) can be misleading when using cross-sectional data.

The Internet and Social Participation: Contrasting Cross-Sectional and Longitudinal Analyses - Irina Shklovski, Robert Kraut, Carnegie Mellon University 
Lee Rainie, Pew Internet & American Life Project 
Full Text: jcmc.indiana.edu/vol10/issue1/shklovski_kraut.html
Abstract: The Internet opens new options for communication and may change the extent to which people use older communication media. Changes in the way people communicate are important, because communication is the mechanism people use to develop and maintain social relationships, so valuable for their physical and mental health. This paper uses data from a national panel survey conducted in 2000 and 2001 to examine the influence of Internet use on communication and on social involvement. In doing so, it contrasts the conclusions one can draw from cross-sectional and longitudinal data on these issues. Longitudinal analyses provide stronger evidence of the causal effects of using the Internet than do the cross-sectional ones. The longitudinal data show that heavy use of the Internet is associated with reductions in the likelihood of visiting family or friends on a randomly selected day. Cross-sectional analyses show high correlations between the frequency with which respondents communicate with specific family members by visits, phone calls and email, suggesting that communication in one medium stimulates the others. In contrast, longitudinal analyses suggest that the links between communication media are asymmetric: visits drive more email communication and phone calls drive more visits, but email drives neither phone calls nor visits.

LONGITUDINAL RESEARCH

Longitudinal studies measure relationships between variables over a period of time. Involving information about an individual or group at different times throughout a long period.

One might follow a group of males from birth to age 30 to measure their involvement with the criminal justice system over time and relate this information to their parents' socio-economic status.

A series of cross-sectional investigations taken over time will provide a longitudinal study.

Recent developments in analytical methods and in data collection activities, including the growing number of longitudinal data sets in Canada and worldwide, have allowed for the increased reliance on powerful longitudinal approaches by research projects.

The Centre for Longitudinal Studies - Following lives from birth and through the adult years. - (CLS) is an ESRC Resource Centre based at the Institution of Education. - cls.ioe.ac.uk

Longitudinal studies of effects of divorce on children in Great Britain and the United States - AJ Cherlin, FF Furstenberg Jr, L Chase-Lansdale, KE Kiernan, PK Robins, DR Morrison, and JO Teitler - Department of Sociology, Johns Hopkins University, Baltimore
National, longitudinal surveys from Great Britain and the United States were used to investigate the effects of divorce on children. In both studies, a subsample of children who were in two-parent families during the initial interview (at age 7 in the British data and at ages 7 to 11 in the U.S. data) were followed through the next interview (at age 11 and ages 11 to 16, respectively). At both time points in the British data, parents and teachers independently rated the children's behavior problems, and the children were given reading and mathematics achievement tests. At both time points in the U.S. data, parents rated the children's behavior problems. Children whose parents divorced or separated between the two time points were compared to children whose families remained intact. For boys, the apparent effect of separation or divorce on behavior problems and achievement at the later time point was sharply reduced by considering behavior problems, achievement levels, and family difficulties that were present at the earlier time point, before any of the families had broken up. For girls, the reduction in the apparent effect of divorce occurred to a lesser but still noticeable extent once preexisting conditions were considered. - sciencemag.org/cgi/content/abstract/252/5011/1386

Legacies and Lessons: Insights from Longitudinal Studies of Educated Women.
Abstract: Postmodern, multicultural, and feminist critiques of psychology have changed how longitudinal researchers construct their inquiries and frame their data. Also the new scholarship brings to the longitudinal investigation perspectives from other disciplines including sociology, anthropology, and history, among others. The book used as a framework for this discussion, "Women's Lives Through Time," is an assembly of different studies, thus providing perspective on the changing discipline of longitudinal research. Many studies on women were out of print and there was an absence of interdisciplinary dialogue about women's adult development. Some broad themes weaving together all the studies are: (1) all of the women studied had attended college; (2) strong evidence existed for increased well-being and feelings of competence in women as they mature; and (3) all the women were affected by role socialization and gender discrimination. The studies included in the book indicated that longitudinal methodologies have diversified over the years. More subjective and context-sensitive methods were incorporated; some caution needs to be taken to ensure that the research retains analysis of measurable data. Another methodological challenge has to do with incorporating issues of socio-historical context into lifespan research. Incorporating subjective and contextual material is the wave of the future in life studies research. - eric.ed.gov

Longitudinal Studies of Attitude Change: Issues and Methods.
Abstract: This report makes available in condensed form the methods for performing longitudinal studies of attitude change and the issues associated with these methods. It represents a state-of-the-art review of such methods, and the material covered spans the disciplines of psychology, sociology, and statistics. It is meant to provide a brief description of the tools available, references to more detailed descriptions, and an overview of the theoretical and practical issues involved in execution and interpretation to those wishing to employ longitudinal methodology. The report is divided into the following four sections: (a) Longitudinal design and cross-sectional design are compared in terms of relative advantages and disadvantages. (b) Theoretical issues associated with studies done over time, such as the overcorrection/the overall variance, are discussed. (c) Actual methods are described; this section is subdivided into (a) experimental designs, (b) quasi-experimental designs, and (c) statistical techniques. This section also includes references to actual studies in which these methods were employed. (d) Practical considerations in doing longitudinal research are discussed. The major conclusion drawn in the report is that while sophisticated methods are available for use, overall development of the field is being hampered because work progresses independently in several disciplines. - eric.ed.gov

Weatherall, R, H Joshi, S Macran. Double burden or double blessing - employment, motherhood and mortality in the Longitudinal-Study of England and Wales. Social Science & Medicine 1994; 38 (2): 285-297.
The OPCS Longitudinal Study has been used to follow up women who were married at the time of the 1971 census, to see if their employment status and responsibility for children at that time had any detectable consequence for their mortality up to 1985. Of particular interest was whether the combination of employment and child rearing produced any signs of role overload, or its opposite hypothesized effect, role enhancement. The results show poorer health among those with neither employment nor children, but these effects did not appear to interact. We suspect the data reveal health selection as much as health effects of the roles taken separately. Whatever the stresses and strains of combining jobs and child rearing, they do not appear drastic enough to result in early death.

How to handle informed consent in longitudinal studies when participants have a limited understanding of the study 
G Helgesson, The Centre for Bioethics at Karolinska Institutet and Uppsala University, Sweden
J Ludvigsson, The Division of Paediatrics at the Department of Molecular and Clinical Medicine, Linköping University, Sweden
U Gustafsson Stolt, The Division of Paediatrics at the Department of Molecular and Clinical Medicine, Linköping University, Sweden 
PhD in philosophy, Researcher at the Centre for Bioethics at Karolinska Institutet and Uppsala University, Sweden; Gert.Helgesson@bioethics.uu.se
Empirical findings from a Swedish longitudinal screening study show that many of the research subjects had a limited understanding of the study. Nevertheless they were satisfied with the understanding they had and found it sufficient for informed continued participation. Were they wrong? In this paper, it is argued that the kind of understanding that is morally required depends partly on the kind of understanding on which the research subjects want to base their decisions, and partly on what kind of knowledge they lack. Researchers must ensure that the information process is not flawed and that participants receive the information they want. To achieve this, new information efforts may be needed. Researchers must also ensure that research subjects have knowledge about aspects of importance to them. Lack of understanding may, however, be the result of conscious choices by research subjects to disregard some of the information because it is not important to them. Such choices should normally be respected. - jme.bmj.com/cgi/content/abstract/31/11/670

Moser, K A, P O Goldblatt. Mortality of Women in the OPCS Longitudinal Study: Differentials by Own Occupation and Household and Housing Characteristics. LS working paper 26. 1985. 

Moser, K A, P Goldblatt. Mortality of Women in Private and Non-private Households Using Data from the OPCS Longitudinal Study. LS working paper 14. 1984. 

Macran, S. Analysis of women's mortality using the OPCS Longitudinal Study. In: OPCS / SSRU Longitudinal Study Newsletter No. 9 (November 1993). 1993. p. 4-7.

Lyons, M. Chaos or complexity? Casualisation, feminisation and gentrification in London, 1971-1991. In: Creeser, R, Gleave, S, editors. Migration Within England and Wales Using the Longitudinal Study. ONS Series LS, No. 9. London: The Stationery Office; 2000. p. 49-61.

Harrop, A, H Joshi. Death and the Saleswoman: an Investigation of Mortality and Occupational Immobility of Women in the Longitudinal Study of England and Wales. LS working paper 73. 1994. 

Goldblatt, P O. Social Class Mortality Differentials of Men Aged 15-64 in 1981: a Note on First Results from the OPCS Longitudinal Study for the period 1981-83. LS working paper 42. Updated version in Population Trends No 51. 1986. 

Fox, A J, D R Jones. 1971-81 Male Socio-Demographic Mortality Differentials from the OPCS Longitudinal Study. LS working paper 21. Shortened version published in Population Trends, 1985, 40: 10-16. Full version published in P O Goldblatt, Proceedings of the American Statistical Association Meeting, August 13-16, 1984. 1984. 

Fielding, A J, S Halford. A longitudinal and regional analysis of gender-specific social and spatial mobilities in England and Wales, 1981-91. In: Boyle, P, Halfacre, K, editors. Migration and Gender. London: Routledge; 1998. 

Fielding, A J. Gender, class and region in England and Wales - a longitudinal analysis. Ritsumeikan University Geographical Journal, Kyoto 1998; 10: 1-22.

Donkin, A. Does living alone damage men's health? Reports on an analysis of the relationship between living alone and risk of death or limiting long-term illness for men who were present in the Office for National Statistics Longitudinal Study (LS) in both 1971 and 1981. Health Statistics Quarterly 2001; 11 (Autumn 2001): 11-16.

Blackwell, L. Women and Science Teaching: the Demographic Squeeze. Report to the DTI. mimeo. Centre for Longitudinal Studies; 2001

E-Books