Chapter 3 The Study of Adult Development and Aging: Research Methods

Content validity:

provides indication of whether a test designed to assess factual material accurately measures the material (e.g., an exam will have questions to see how well you understand topics covered in this chapter)

Because age cannot be experimentally manipulated, studies of aging represent a

quasi-experimental design Researchers compare groups on predetermined characteristics (e.g., age, sex, ethnicity, SES). Cannot conclude that age caused variations in the dependent variable, but can use results to describe differences between age groups.

Prospective study:

A variant of longitudinal design in which researchers sample from a population of interest before they develop a particular type of illness or experience a particular type of life event E.g., study widowhood by examining a sample of married individuals (assuming that a certain % will suffer death of spouse during time of study)

Longitudinal Design

A research design where people are followed repeatedly from one test occasion to another. Researchers attempt to determine whether participants have changed over time as a result of the aging process

Validity

A test's ability to measure what it is supposed to measure

Criterion validity:

indicates whether a test score accurately predicts performance on an indicator measure (GRE scores predict success in grad school).

Internal consistency:

indicates whether respondents answer similarly on comparable items.

Cross Sectional Designs

A research design where groups of people are compared with different ages at one point in time In studies of aging, typically older adults 65+ are compared with younger adults (e.g., of college age). More frequently used research design

Descriptive Research Designs

A design that provides information about age differences but does not attempt to rule out social or historical factors. Example: Study shows that older adults more likely to have certain health problems than are younger adults Is aging what results in older adults having certain health problems? Cannot determine whether a health problem is a natural result of aging or result of environmental conditions present during the individual's lifetime. Findings may reflect differences in early life experiences instead of age.

Reliability

A measure's consistency and whether it produces the same results each time it is used.

Correlational Designs

Alternative approach to describing group differences A research design in which researchers investigate relationships among two or more variables. Relationships are observed among variables as they exist in the world. Participants are not divided into groups nor are variables manipulated.

Age is not a true independent variable (although developmental psychologists classify age as an independent variable) Why?

Because age cannot be manipulated or controlled. Studies on aging are never true experiments Cannot establish that aging causes people to receive certain scores on a dependent variable Can establish whether different age groups vary in responses to the experimental manipulation.

Longitudinal Designs Strengths

Can add valuable data on psychological changes in adulthood and old age. As data accumulates from multiple studies on related variables, a body of evidence can build up that helps inform larger research questions.

Longitudinal Designs Limitations

Cannot differentiate between aging within the individual from changes in social and historical context. Expensive and logistically difficult Final results take many years Practice effects Tests become outdated

Variable:

Characteristic that varies from individual to individual

What is the main limitation of a descriptive research design?

Developmental researchers are interested in the impact of chronological age on a person's functioning. Problem with this - Need to disentangle a person's age from their historical period cannot resolve the problem because they do not allow researchers to manipulate age and environmental influences in a way as to disentangle them.

Simple Correlational Designs

Examine the relationship between two variables Age is treated as a continuous variable and its relationship with another variable is expressed through a correlation coefficient ("r") Can be negative or positive Positive correlation: as one variable increases the other variable increases Negative correlation: as one variable increases the other variable decreases Strength is indicated by how close it is to 1 (or -1). Limitation: No ability to detect causality.the researcher cannot make any assumption about what caused what - there are no independent or dependent variables.

Independent variable

In an experimental study, the factor that the experimenter manipulates

Dependent variable

In an experimental study, the factor that the experimenter observes

Case reports

In-depth analysis of particular individuals Data may be integrated from interviews, psychological tests, observations, archival records, journals and diary entries. Focus is on the characteristics of the individual and what has influenced his/her development and life experiences Personal narratives also obtained where individuals describe their lives as they experienced them along with their ideas about why their lives evolved in a given manner Advantages - Provides insight into the lives of individuals as they change over time. Disadvantages - Relies heavily on clinical judgement by researcher

Multivariate Correlational Designs

Involves analysis of relationships among more than two variables (in contrast to a simple correlational design) Two or more variables can be examined at a time (to control for extraneous factors that might affect relationship between variable of interest on dependent variable)

Longitudinal Designs Limitations (cont.) Attrition -

Participants may die or drop out of the study More difficult to make sense of data due to few participants left to allow for stat analyses across multiple time points. Worsened by Selective attrition - people who drop out of a longitudinal study may not represent sample originally tested (results in skewed data) To address problem: conduct analyses to determine whether pattern of dropout is random or due to systematic bias Non random sampling - Successive samples are increasingly unlike populations they were intended to represent

Types of Research Methods Laboratory studies

Participants tested in systematic fashion using standardized procedures, often involving some type of task. Majority of info about physical and cognitive changes associated with aging comes from lab studies. Advantages: objective and systematic way in which data is recorded provides researcher assurance that results are due to variables studied than to extraneous factors. Disadvantages: inability to apply the stimuli presented to real life experiences.

Cross Sectional Designs Strengths

Quick and inexpensive Latest technologies can be used Can use statistical controls to ensure differences other than age kept to a minimum. Findings regarded as tentative descriptions of effects of aging that can be replicated through longitudinal studies.

Experimental Design:

Research method in which an independent variable is manipulated and scores are then measured on the dependent variable. Involves random assignment of research participants to treatment and control groups Each group represents different levels of the independent variable. Performance of groups compared on the dependent variable. If groups differ on dependent variable, researcher assumes it is due to exposure to different levels of the independent variable.

Observational method

Researchers draw conclusions about behavior through careful and systematic examination in particular settings. One type is Participant observation: researcher participates in activities of the respondents. Recordings may be made using video tapes or behavior records Elaborate procedures available for creating behavioral records where researcher precisely defines behavior to be observed and specifies the times during which records will be made. Procedure may be used to determine whether an intervention is having its intended effects by comparing behavioral records before and after intervention. *advantages/limitation ?

Cross Sectional Designs Limitations

Results reflect age differences not age changes Studies need to control as much as possible for cohort effects that can obscure effects of age Select younger samples comparable in important ways to the older sample All participants are studied at one time in history Results reflect not only differences between cohorts but also effects of social and cultural influences.

Cross Sectional Designs Limitations (cont.) Selective survival

Study participants are survivors of their respective age groups. Represent healthier [or luckier] group of people than those in their cohort who did not live long Consequently, older adults may look different from younger ones because the two groups are drawn from two different populations Participants who die young but still represented in young adult group Participants who survive to be old

Discriminant validity:

demonstrates that the measure does not relate to other measures that have no theoretical relationship (scores on a depression measure should not correlate with a measure of ADHD sxs)

Test-retest reliability:

determined by giving a test on two occasions to assess whether respondents receive similar scores across both administrations

Age

how many years person has lived measurement of change within the individual

Convergent validity:

needed to determine if the measure relates to other measures that are theoretically similar (do scores on a depression measure correlate with scores on another measure of depression?).

time of measurement effect

social, historical, and culture influence that are presently affecting people participating in developmental research

cohort effect

the social, historical , culture influence that affect people during a particular period of time

Construct validity:

used to assess extent to which a measure is intended to assess psychological constructs is able to do so

time of measurement

year or period in which a person is tested in research measurement of current influences on individuals being tested

corhort

year or period of a person's birth measurement of influence relative to history of birth