Event history analysis helps the researcher to understand the reasons behind why certain individuals are exposed to a higher risk of experiencing certain events of interest than other individuals. Event history analysis can be accomplished by using special types of methods.
Survival analysis consists of a wide variety of methods or techniques that help the researcher to analyze the timing of events.
The first step in event history analysis is for the researcher to understand the nature of the event history data and the reason why the event history analysis is being done. In other words, one can say that in event history analysis, the researcher defines the terms and states that one wishes to distinguish. For example, suppose in an event history analysis of marital histories, four types of states are generally obtained. The four states in this type of event history analysis are: ‘never married,’ ‘married,’ ‘divorced,’ and ‘widowed.’ The term ‘event’ in event history analysis is nothing but a transition from one state to another. In this example of event history analysis, the event can be referred to as ‘married,’ which is a transition from the origin of ‘never married’ to the final state of ‘married.’
Another important concept in event history analysis is that of risk period. It is a basic fact that while conducting event history analysis, the researcher can never encounter a case where the person has experienced each of the events under study. In event history analysis, the risk period plays a role in a case when a person is exposed to the given risk under consideration. For example, in event history analysis, considering the above example, a 28 year old never married person is exposed to a greater “risk” of getting married. Thus, the period that someone is at risk of a particular event is the risk period in event history analysis.
With the help of such concepts, one can define event history analysis as the analysis that involves the study of the duration of the non-occurrence of an event during the risk period. In event history analysis, the dependent variable is not the time or duration, but the transition rate. There is another definition of event history analysis which is the analysis of the degree or rate of occurrence of the event during the risk period.
The most typical type of event analyzed in survival analysis is that of death. Survival analysis can also be used to analyze other events as well, like unemployment, graduation from school, etc.
The survival time data in survival analysis has two important special characteristics. The survival time data in survival analysis is not negative and is usually positively skewed.
Some subjects of survival time data in survival analysis are not observed. In other words, the event of interest does not take place for the subjects during the study in survival analysis.
The survival time, which is the object of study in survival analysis, should be differentiated from the calendar time. The survival time in survival analysis should always be measured related to some appropriate time origin.
Survival analysis can be used to study many things. It is extremely helpful in studying the cause of births and deaths. Survival analysis can also be helpful if the researcher wants to study the cause of job changes and promotions. Further, survival analysis is used by the researcher in order to understand the cause of marriages and divorces. Survival analysis can also be used by the researcher to study the cause of residence changes or the causes behind wars and revolutions.
Contact Statistics Solutions today for a free consultation on Event History and Survival Analysis.
