Using Survival Analysis in Research
This easy tutorial will show you how to run the Survival Analysis Test in SPSS, and how to interpret the result.
Survival analysis is one of the advanced research methods used in many different sciences. That is to say, the name of this method reflects its beginnings in medical and demographic studies of mortality. In addition, following the introduction of the proportional hazard model by Cox in 1972, empirical analysis of event history data has become more popular and has become even more widespread.
Survival analysis is a part of statistics that deals with the analysis of time until an event occurs. To clarify, the subject of observation is a variable that having positive values, such as time in death in medicine, length of stay in the hospital, time to recovery when receiving some therapy, or in the economy strike duration, unemployment, etc.
Assumptions for Survival Analysis
When performing a Survival Analysis Test procedure the following assumptions:
- event status: censored or event (two mutually exclusive states).
- survival time – time to an event or censorship.
- independence of censoring and the event.
- a similar amount and pattern of censorship per group.
An Example: Survival Analysis Test in SPSS
This guide will explain, step by step, how to run Survival analysis Test in SPSS statistical software by using an example.
We use the data from Luke, D.A., & Homan, S.M. (1998). Researchers wanted to examine factors associated with alcohol relapse. Therefore, they had two groups of people:
the first group was people who were submitted to detoxification, and the second group was people who were submitted to another treatment.
Finally, we define the survival time (time to relapse) as the number of days from detoxification or hospitalization until the onset of drinking.
For a Kaplan-Meier survival analysis, we need at least for variables:
- ID – case identifier; participant;
- Survival time – For example, defined in weeks;
- Event status – For example defined as 1 – event and 0 – censored;
- The between-subjects factor – for example group; consists of two groups that we are comparing; 0 – detoxification and 1 – treatment.
This easy tutorial will show you how to run the Survival Analysis Test in SPSS, and how to interpret the result.
