CHAPTER 2
THEORY AND METHOD IN STUDYING AGING AND OLDER ADULTHOOD
CHAPTER SUMMARY
Chapter 2 is divided into five main sections. The first section
... [Show More] describes three viewpoints, or metamodels, that guide developmental research – mechanistic, organismic, and contextual. Each viewpoint has its own way of looking at the role played by the organism and by the environment during development. Then the key propositions of the lifespan developmental perspective are listed, followed by a discussion of the proportion of developmental gains and losses over the life span.
The second section of the chapter on developmental research discusses the age variable, which is organismic, and the issue of age boundaries when an older age group is included in a research study. Three basic factors must be considered when conducting developmental research: age, cohort, and time of measurement. The latter two factors are discussed in detail and examples are given.
Three research designs are described. The cross-sectional design, the most commonly used and most efficient, confounds age and cohort. The longitudinal design follows individuals over time and confounds age and time of measurement. The time-lag design focuses on only one chronological age group and confounds cohort and time of measurement. Schaie’s sequential research designs attempt to disentangle confounded factors and determine which ones are significant.
The third section of the chapter covers measurement issues. Reliability refers to the dependability of research instruments as well as the consistency of results obtained in research studies. Once reliability is established, the concern is with validity. Validity has to do with whether we are measuring what we think we are measuring. There are several kinds of validity (internal, external, ecological) and each is described. Other issues discussed are the meaning of heterotypic continuity and the importance of sampling for the validity of a research study.
The fourth section covers approaches to conducting aging research. With the experimental approach, studies have independent and dependent variables, study participants are randomly assigned to different levels of the manipulated independent variable, the outcome is measured by the dependent variable, and cause-and-effect statements can be made. Quasi-experimental studies appear to have the same form as experimental studies, but study participants are not randomly assigned to levels of an “independent” variable, so cause-and-effect statements cannot be made. An example is given for how each approach can be used to study activity level and memory in older adults. Experimental and quasi-experimental approaches can have single-factor or multi-factor designs. A third, descriptive, approach involves collecting data on variables of interest and measuring the relationship between them. This relationship is computed statistically using correlations, which range from -1.0 to 0 to +1.0 [Show Less]