Chapter 3: Research Methods Flashcards
Quasi-experimental design
researchers compare groups on predetermined conditions (e.g. age) but cannot conclude causation
Descriptive (single-factor) designs
studies that catalogue information about how people perform based on their age but do not attempt to rule out social or historical factors (e.g. longitudinal and cross-sectional designs)
Cohort
the year or period of a person’s birth
Cohort effects
social, historical, and cultural influences that affect people during a particular period in time; normative history-graded influences present around the time of a person’s birth
Time of measurement effects
social, historical, and cultural influences that are presently affecting participants in a study; also normative history-graded influences
Longitudinal design
data is collected from the same people repeatedly over several decades
Prospective study
kind of longitudinal design where researchers sample from a population of interest and compare data from before and after a particular type of illness or life event to examine risk/protective factors
Limitation of longitudinal studies
unable to differentiate between aging within the individual and changes in the social and historical context
Selective attrition
special case of non-random sampling when a study loses participants, causing the sample to be unrepresentative of the original one and the data to be skewed
Terminal decline
individuals gradually lose cognitive abilities as they approach death
Cross-sectional designs
researchers compare groups of people with different ages at one point in time; a snapshot in time
Sequential designs
a sequence of studies (e.g. cross-sectional) that are carried out over years; can keep adding cohorts which can address atttrition
Most efficient design
a set of 3 designs manipulating the variables of age, cohort, and time of measurement: time-sequential, cohort-sequential, and cross-sequential designs
Time-sequential design
examines the effects of time of measurement in contrast to age
Cohort-sequential design
cohorts are compared at different ages
Cross-sequential design
cohorts are examined at different times of measurement
Correlational design
relationships are observed among variables as they exist in the world with no attempt to divide participants into groups or to manipulate variables; age is treated as a continuous variable
Types of studies
laboratory studies, qualitative studies, archival research, surveys, epidemiological studies, case reports, focus groups, daily diaries, observational methods, meta-analysis
Laboratory studies
participants are tested in a systematic fashion using standardized procedures; most objective
Archival research
researchers use existing resources that contain data relevant to a question about aging (e.g. newspapers)
Epidemiology
study of the distribution and determinants of health-related states or events, and its application to the control of diseases and other health issues
2 types of population estimates provided by epidemiological studies
prevalence and incidence statistics
Prevalence statistics
provides estimates of the percentage of people who have ever had symptoms in a particular period
Incidence statistics
provides estimates of the percentage of people who first develop symptoms in a given period
Case reports
summarizes findings from multiple sources for an in-depth analysis of particular individuals
Focus group
a meeting of a group of respondents oriented around a particular topic of interest; often a pilot study
Daily diary method
participants provide data on a daily basis for researchers to examine day-to-day changes
Observational method
careful and systematic examination of behavior in particular settings
Meta-analysis
statistical procedure that allows researchers to combine findings from independently conducted studies on similar phenomena by calculating an effect size
Pros of quasi-experiments
allows researchers to examine effect of a treatment that may not be ethical or logistically possible in an experiment
Cons of quasi-experiments
have less internal validity than experiments (no random assignment) and unable to determine if aging causes changes
(Bivariate) Correlation (r)
statistic that indexes the degree or strength of a relationship between 2 continuous variables
Cons of correlation
cannot infer causation, only useful for linear relationships, unable to determine if there are cohort or time of measurement effects
Age effects
any differences caused by underlying processes (like biological or psychological) that occur with aging
