Human memory and many of ideas and theories about it has been a subject of science and philosophy for thousands of years and has become one of the major topics of interest within cognitive psychology and a great concern among scientists.
This is because; human memory differs from one person to another. In addition, the person's memory may behave in different way according to many aspects such as, time, age, sex, … , etc.
Therefore, there was a great importance to establish models for measuring human memory. In this thesis, we are interested in using a statistical mixture model for measuring the memory effects and response error due to time. The effects of time and memory factors on response can be described by a function that is the product of effects due to omission effects and telescoping effects.
Several authors have considered the effect of time on the respondent's memory by setting models for memory effects, see, e.g., Mahalanobis (1946), Gray (1955), Mooney (1962), Neter and Waksberg (1964), Sudman and Ferber (1970), Som (1973), Sudman and Bradburn (1974), Moss and Goldstein (1979), and Dirk Sikkel (1985), among others.
Mixture models have been applied in many fields and used to analyze and model measurements for many experiments. The mixture model that will be presented in this thesis can give a good insight into the structure of memory effects. Many scientists have used mixture models, see, e.g., Cohen (1960), Blischke (1964), Ashton (1971), Hosmer( 1973), John Wiley & Sons (1985), Airoldi, Flury & Biber (1992), Pettier Nurmi (2001), among others.
|
