Design assessing age-related changes; data are obtained from the same individuals over a long period

Longitudinal studies differ from one-off, or cross-sectional, studies. The main difference is that cross-sectional studies interview a fresh sample of people each time they are carried out, whereas longitudinal studies follow the same sample of people over time.

Features of longitudinal vs cross-sectional studies

Download this table as a PDF handout

Some cross-sectional studies take place regularly, each time including a large number of repeat questions. For example, the British Social Attitudes Survey is a repeat cross-sectional study that has been carried out nearly every year since 1983. It provides excellent data about how Britain’s attitudes and values have changed (or not changed) over time.

Repeating the same questions in each round allows researchers to look at how society as a whole has changed over time. But because the questions are asked of a new sample every time, these studies can only reveal change at an aggregate level – they can shed little light on who has changed, or how or why.

For example, the 2015 British Social Attitudes (PDF) survey found that 66 per cent of people thought that “it’s everybody’s duty to vote” in a general election, down from 76 per cent in 1987.

What do these findings tell us? The data clearly show us that, overall, fewer people now than in the late-1980s think that citizens have a duty to vote. We can look at the characteristics of those who do or don’t agree with this view, and how the profile of these groups had changed over time. We can also examine how the likelihood of thinking that voting is a duty has changed among different population groups (for example, different age groups or ethnicities).

These sorts of calculations would provide some very helpful insights. But there are many things that this kind of cross-sectional data cannot tell us, but which longitudinal data would help us to address. For example:

• Which individuals changed their views about voting over the period? What are their characteristics? Do some people switch, and then switch back again?
• The British Social Attitudes findings show us the net or aggregate change over time – the difference between the proportion who thought voting was a duty in 1987 and the equivalent proportion now. But this tells us nothing about change at an individual level. For example, what proportion became more inclined to think voting is a duty and what proportion became less inclined? It is important to bear in mind that if these two proportions are similar, they could cancel one another out at a net level and make it appear that very little has changed over all, even though substantial numbers of people have changed their views.
• What factors best explain the transition away from thinking voting is a duty?

Sometimes data from longitudinal studies is analysed cross-sectionally. This means that the researcher is just focusing on the information collected at one round of the study, and not linking that information to data from earlier or later rounds.

Study design depends greatly on the nature of the research question. In other words, knowing what kind of information the study should collect is a first step in determining how the study will be carried out (also known as the methodology).

Let’s say we want to investigate the relationship between daily walking and cholesterol levels in the body. One of the first things we’d have to determine is the type of study that will tell us the most about that relationship. Do we want to compare cholesterol levels among different populations of walkers and non-walkers at the same point in time? Or, do we want to measure cholesterol levels in a single population of daily walkers over an extended period of time?

The first approach is typical of a cross-sectional study. The second requires a longitudinal study. To make our choice, we need to know more about the benefits and purpose of each study type.

Cross-sectional study

Both the cross-sectional and the longitudinal studies are observational studies. This means that researchers record information about their subjects without manipulating the study environment. In our study, we would simply measure the cholesterol levels of daily walkers and non-walkers along with any other characteristics that might be of interest to us. We would not influence non-walkers to take up that activity, or advise daily walkers to modify their behaviour. In short, we’d try not to interfere.

The defining feature of a cross-sectional study is that it can compare different population groups at a single point in time. Think of it in terms of taking a snapshot. Findings are drawn from whatever fits into the frame.

To return to our example, we might choose to measure cholesterol levels in daily walkers across two age groups, over 40 and under 40, and compare these to cholesterol levels among non-walkers in the same age groups. We might even create subgroups for gender. However, we would not consider past or future cholesterol levels, for these would fall outside the frame. We would look only at cholesterol levels at one point in time.

The benefit of a cross-sectional study design is that it allows researchers to compare many different variables at the same time. We could, for example, look at age, gender, income and educational level in relation to walking and cholesterol levels, with little or no additional cost.

However, cross-sectional studies may not provide definite information about cause-and-effect relationships. This is because such studies offer a snapshot of a single moment in time; they do not consider what happens before or after the snapshot is taken. Therefore, we can’t know for sure if our daily walkers had low cholesterol levels before taking up their exercise regimes, or if the behaviour of daily walking helped to reduce cholesterol levels that previously were high.

Longitudinal study

A longitudinal study, like a cross-sectional one, is observational. So, once again, researchers do not interfere with their subjects. However, in a longitudinal study, researchers conduct several observations of the same subjects over a period of time, sometimes lasting many years.

The benefit of a longitudinal study is that researchers are able to detect developments or changes in the characteristics of the target population at both the group and the individual level. The key here is that longitudinal studies extend beyond a single moment in time. As a result, they can establish sequences of events.

To return to our example, we might choose to look at the change in cholesterol levels among women over 40 who walk daily for a period of 20 years. The longitudinal study design would account for cholesterol levels at the onset of a walking regime and as the walking behaviour continued over time. Therefore, a longitudinal study is more likely to suggest cause-and-effect relationships than a cross-sectional study by virtue of its scope.

In general, the research should drive the design. But sometimes, the progression of the research helps determine which design is most appropriate. Cross-sectional studies can be done more quickly than longitudinal studies. That’s why researchers might start with a cross-sectional study to first establish whether there are links or associations between certain variables. Then they would set up a longitudinal study to study cause and effect.

Source: At Work, Issue 81, Summer 2015: Institute for Work & Health, Toronto

This column updates a previous column describing the same term, originally published in 2009.

What research method is used to assess age related changes in which data is obtained from the same individuals at intervals over a long period of time?

What is an experimental design for assessing age related changes in which data are obtained simultaneously from people of differing ages?

What type of research design assesses multiple cohorts over time?

What is longitudinal design in research methods?