Natural experiment

From Wikipedia, the free encyclopedia

A natural or Quasi-experiment is a naturally occurring instance of observable phenomena which approach or duplicate a scientific experiment. In contrast to laboratory experiments, these events aren't created by scientists, but yield scientific data. Natural experiments are a common research tool in fields where artificial experimentation is difficult, such as Cosmology, epidemiology, and sociology.

An example of a natural experiment occurred in Helena, Montana during the period from June 2002 to December 2002 when a smoking ban was in effect in all public spaces in Helena including bars and restaurants. Helena is geographically isolated and served by only one hospital. It was observed that the rate of heart attacks dropped by 60% while the smoking ban was in effect. ^[1]

A further example is the spread of early humans across the Pacific Ocean. The distribution of populations between islands was essentially random, allowing researchers to treat different groups as independent societies drawn from a common pool. Hypotheses can then be tested in different contexts without fear that an unobserved factor is the cause of differences between island populations. This procedure would be inappropriate in studying European colonies, which were typically planned by the metropolitan government and staffed by carefully-chosen populations.

An example of a Quasi-Experiment is as follows:

Quasi-Experiment

Methods

The purpose of this proposed study is to analyze the effects that an insignificant positive event has on a person's subsequent mood by observing reactions of subjects after receiving two sodas from a vending machine after only paying for one. In order to assess such effects we will conduct an observational study with intervention. Our study will follow a quasi-experimental design, as we will not use random selection to choose participants. Instead, participants will be chosen as a result of their own decision to use a specific vending machine. The independent variable in this study is the number of sodas the participant receives, which we will manipulate by rewiring the vending machine. The dependent variable, which we plan to analyze, is the mood of the participant after receiving the sodas. Specifically, we will measure the difference in mood between the time when the participant approaches the vending machine and when he or she leaves the machine. We hypothesize that after receiving a second soda the participant will have an increase in positive mood. Thus, we expect that a positive insignificant event will have a significant and immediate positive effect on one's mood.

Participants

Fifty participants, sex unknown, will participate in this study. No person will be picked or assigned to a group. The vending machine will be controlled to release an extra soda exactly every-other time that data is being collected. Exactly twenty-five participants will receive an extra soda. Participants will be excluded from the study if there is a group present at the vending machine. Data will be collected and an extra soda will only be released in situations where one individual is present at the vending machine.

Procedure

This was a quasi-experimental design, with one independent variable and one dependent variable. Quasi-experimental designs are similar to experimental designs; however, there is no random assignment of participants. The independent variable is the number of sodas released from the soda machine, alternating between one and two. The dependent variable in this study is the mood of the buyer. The participants’ moods were evaluated as they approached the machine, as happy, neutral, or unhappy. Happy was classified as animated, smiling or laughing, neutral as expressionless, and unhappy as stressed, frowning or crying. An increase in positive feelings was noted if the participant smiled or laughed (more than before they received the drinks if they were considered happy to begin with) or if they mentioned their luck to someone nearby. An increase in happiness was also noted if an unhappy person seemed to forget their unhappiness and relax, smile, or stop crying.An increase in negative feelings was measured by any decrease in classification: happy people becoming neutral, happy people becoming unhappy, neutral people becoming unhappy, or unhappy people becoming even more unhappy then previously. If the participant experienced neither an increase in happiness nor unhappiness after receiving the sodas their mood was classified as consistent. In conducting the experiment, three researchers sat near a soda machine rigged to release two sodas when the buyer only paid for one every other time. If a group of two or more people approached the machine they were not counted in the study. The soda machine was in a public place with benches nearby so the researchers did not raise suspicion and did not affect the dependent variable (the participant’s mood). The researchers recorded gender and an increase, decrease or no change in the buyer’s mood for a duration of 50 trials.

[edit] References

1. ^ Sargent RP, Shepard RM, Glantz SA, 'Reduced incidence of admissions for myocardial infarction associated with public smoking ban: before and after study.', British Medical Journal, volume 328 pp. 977-980, 2004