The motivation for developing Program MARK was to bring a common interface to the estimation of survival from marked animals. Marked animals can be re-encountered as either alive or dead, in a variety of experimental frameworks. Prior to MARK, no program combined the estimation of survival from both live and dead re-encounters.
Program MARK provides survival estimates from marked animals when they are re-encountered at a later time. The time intervals between re-encounters do not have to be equal. If not specified by the user, these time intervals are assumed to be 1. More than one group of animals can be modeled, e.g., treatment and control animals. The re-encounters can be from dead recoveries (e.g., the animal is harvested), from live recaptures (e.g. the animal is re-trapped), or from radio-tracking. See Model Structure for more detail.
The basic input to Program MARK is the encounter history for each animal. The user can use a summarized form of the encounter histories when multiple animals have the same history. See Encounter Histories File for more details of how encounter histories are formatted for use by MARK.
In addition to providing the survival estimates, these estimates can be developed as part of a model. That is, an underlying model is built that predicts the survival estimates. Parameters can be constrained to be the same across re-encounter occasions, or by age, or by group, using the parameter index matrix, or PIM. See Parameter Matrices for more detail. A set of common models for screening data initially are provided, with time effects, group effects, time*group effects, and a null model of none of the above provided for each parameter, with all possible combinations across all parameters. See Pre-Defined Models for more details.
The model can include covariates, specific to the individual animal or to the re-encounter occasion. Examples are the weight of the animal at the time of marking, or the average temperature over the survival interval. See Design Matrix for more detail. Besides the logit link function, other link functions include the log-log, complementary log-log, sin, log, and identity.
Program MARK computes the estimates of model parameters via numerical maximum likelihood techniques. The FORTRAN program that does this computation also determines numerically the number of parameters that are identifiable in the model, and reports its guess of one parameter that is not identifiable. The number of estimable parameters is used to compute the QAICc for the model.
Outputs for various models that the user has built are stored in a database, known as the Results Database. The input data are also stored in this database, making it a complete description of the model building process. The database is viewed in a Results Browser window.
Summaries available from the Results Browser window include viewing and printing model output (estimates, standard errors, and goodness-of-fit tests), deviance residuals from the model (including graphics and point and click capability to view the encounter history responsible for a particular residual), likelihood ratio and ANODEV between models, and adjustments for over dispersion.
These capabilities are implemented in a Microsoft Windows style interface. Thus, most PC users are familiar with the operation of such programs, and find navigation reasonably easy. Windows can be closed by clicking on the window’s close button, and generally manipulated using standard Window’s conventions. Context-sensitive help screens are available with Help click buttons and the F1 key. The Shift-F1 key can also be used to investigate the function of a particular control or menu item. Help screens include hypertext links to other help screens, with the intent to provide all the necessary program documentation on-line with Help screens.