One of the goals of health psychology is to understand health problems and why they occur.  In order to do this, they want to use research methods that are naturalistic as possible.  If a study is too artificial, then it may not be applied successfully in order to improve a community’s general health. Two naturalistic approaches to research are case studies and natural experiments.

A natural experiment is an experiment in which the independent variable is not manipulated by the researcher, but is naturally occurring.  The researcher takes advantage of something that is happening, but loses the ability to establish a clear cause and effect relationship.  One example is a study done by Kiecolt-Glaser on the effects of stress on the immune system.  In their study, they measured university students’ level of stress one month before exams. As a result of this the participants were identified as either high stress or low stress. A blood sample was taken to measure the level of NK cells, which are a measure of the strength of one’s immune system.  Another blood sample was taken after the students had taken their first two exams.

The researchers found a significant decrease in NK cell count in the high stress group after the second exam.  It appears that the stress experienced during exams lowered their immune functioning. The fact that the researchers measured NK cell levels both before and after the exams (the naturally occurring IV) allows them to draw conclusions about the role of stress on the immune system.  However, since the IV was not manipulated and the study was not highly controlled, it is not possible to draw a true cause and effect relationship.  In addition, the researchers measured the level of loneliness for the students and found that those that had high stress but were less lonely had higher levels of NK cells, indicating that social support may be an important protective factor.

Cohen also carried out a study to see if stress affects one's health - specifically, if high levels of stress leads to colds – but in this case, the researcher used a true experiment in which an independent variable was manipulated, which means that a cause and effective relationship can be established. He started off by measuring people’s level of stress by giving them a questionnaire. He then had them inhale a cold virus or a placebo (saline solution). Participants were then quarantined for five days. Cohen found that those who had high levels of stress for a month before exposure to the virus were twice as likely to get ill.  Unlike the natural experiment, this infers a true cause and effect relationship.  The problem is that it is rather artificial, inhaling a virus directly and being highly controlled.  However, in research high internal validity often means sacrificing ecological validity. Like the study by Kiecolt-Glaser, both of the studies have a limitation of how they measure stress.  Both are based on self-reported questionnaires.  Although it is likely that the participants were honest in their assessment of the stressful experiences they have had over the past six months, it is not a very precise measure of one's level of stress. 

A final research method used in health psychology is a case study.  A case study is a more holistic approach to studying an individual or a group – and it is longitudinal. It is carried out in a naturalistic setting so has high ecological validity – but since it is longitudinal and naturalistic, it is not only low in internal validity, it is difficult to replicate.  Since replication is difficult, it is important that such studies have high sample sizes.  One such study was carried out by Marmot et al.  In the “Whitehall study,” the researchers wanted to see if one’s position in a social hierarchy had an effect on their health.  In addition to questionnaires about their level of stress, researchers used medical records and interviews over a period of five years to see the effects of stress on their cardiovascular health.  Marmot found that the people who were lower in the bureaucratic hierarchy had more health problems than those higher in the system, even when matched for other variables such as smoking, family history of heart disease and level of exercise.  The study is important because it challenged the idea that those at the top of the socioeconomic ladder had the most stress and were most in danger of health problems.  A case study uses method triangulation, which leads to a higher level of credibility.  If the different research methods lead to the same conclusions, then there is a higher level of credibility.

Although using naturalistic research methods is important in health research, they are difficult to replicate and they have low internal validity – meaning that confounding variables may play a role in the findings.  It is important that findings from research in the lab, using controlled experiments, can confirm the findings of natural experiments and case studies.  When this happens, researchers can draw firmer conclusions about cause and effect.