Chapter 3

The Study of Adult Development and Aging: Research Methods

Lecture guidelines

This chapter is the third and final chapter that serves as background for what is to follow in the rest of the course. Given that the topic is not one that inherently of great interest to students, some imagination is required to keep students motivated. However, the slides provide a great deal of guidance to structure the presentation and offer opportunities for discussion. In the slides provided with the text, examples from Rochester Adult Longitudinal Study are provided, but others could be substituted from the instructor’s own data or other published studies.

Videos and films

As stated above, it is helpful to use some imagination in planning the research methods lecture because the material is not of strong inherent interest to students. Suggested videos would include recent studies from the national news media using cross-sectional or longitudinal designs. A segment from “56 Up (http://www.pbs.org/pov/56up/#.UbedgPm1GCl) is perfect for illustrating longitudinal research. “Neil” is a particularly interesting case. The 2019 update (“63 Up”) may also be available to show in class when it is released on video (https://www.youtube.com/watch?v=Ugw7hUoJ0C0).

Discussion questions:

What examples can you come up with of studies reported in the media about adult development and aging, and how would you critique these?

Variables in developmental research

Can you define independent and dependent variables? Think of examples from research and label the variables as independent and dependent.

Can you think of examples of an experimental study? Are studies on aging experimental or quasi-experimental?

Why is it that cohort, time of measurement, and age cannot truly be separated from each other?

Can you give some examples of cohort and time of measurement effects?

Single-factor research designs

Can you explain the problem of attrition in longitudinal studies in your own words?

Which principle in adult development and aging is illustrated by the problem of attrition?

Under what circumstances would a cross-sectional study particularly make sense?

How could you overcome some of the problems in interpreting cross-sectional studies in adult development and aging?

What examples can you come up with of studies reported in the media about adult development and aging, and how would you critique these?

Sequential designs

Can you provide examples of cases where sequential designs would be important in the study of aging?

Which designs seem to make more sense to you and would you use in conducting your own research?

Can you explain, in your own words, what researchers would do in order to separate age, cohort, and time of measurement in sequential studies?

Correlational designs

Why do researchers maintain that it is not possible to draw cause-and-effect conclusions from correlational research?

What types of variables and problems are best studied through correlational methods?

What are the advantages to using multivariate designs in research on adult development and aging?

Types of research methods

How might an older adult feel when tested in a laboratory situation?

For what type of research questions would qualitative methods be particularly well-suited?

In what areas would archival research be of particular value? Why?

What types of special considerations might be needed in conducting surveys on older adults?

What can researchers do to reduce the subjectivity involved in a case report?

How could a focus group help a researcher begin to identify issues that can then be addressed in subsequent empirical studies?

If you were a consultant in a nursing home, how might you use observational methods to answer questions about the factors that affect the adjustment of residents?

Measurement issues in adult development and aging

Why is it necessary to establish the appropriateness of measures for different age groups of adults?

What is the difference between reliability and validity?

What are the three different types of validity, and how can researchers establish these in a given measure used to study adult development and aging?

Ethical issues in research

Why is it necessary to protect the rights of research subjects?

How can researchers protect the rights of older adults who may be unable to provide their own informed consent?

How can debriefing become problematic when it is necessary to inform participants of a study’s true purposes?

Under what sorts of conditions would it be necessary to keep subjects unaware of the purpose of the study until it had been completed? What steps should be followed in such conditions to ensure that subjects are protected?

Multiple choice questions

AgeFeed

As stated in AgeFeed, the _______________ Study of the Gifted and Talented was started almost 100 years ago and is still running.

1. Terman
2. Berkeley
3. Baltes
4. Florida

AgeFeed

The number of supercentenarians being tracked by the Gerontology Research Group is:

1. 110
2. 36
3. 90
4. 1450

AgeFeed

Which species of animal does not get cancer, and is therefore being studied to understand genetic causes of the disease?

1. whale
2. dolphin
3. shark
4. tuna

Variables in developmental research

Random assignment

Difficulty: Hard

A pharmaceutical company conducted an experimental study to investigate the effectiveness of a new medication to treat Alzheimer’s Disease. The people in the group receiving the strongest dose of the drug also had the lowest genetic vulnerability for the disease. Based on this information, which is a reasonable criticism of this study?

1. There was no random assignment of participants to groups.
2. Participants did not provide informed consent.
3. Family members were not consulted about the study.
4. Genetic vulnerability was not relevant to the study.

Age as variable

Because age can never be manipulated, studies on aging can never be considered:

1. experimental
2. valid
3. correlational
4. lab research

Quasi-experimental

The element missing from a quasi-experimental design that makes it impossible to draw cause-and-effect conclusions is:

1. dependent variable
2. hypothesis
3. statistical analysis
4. random assignment

Cohort effect

Which of the following is likely to be considered a "cohort" or "time of birth" effect on psychological performance?

1. a reduction in the health care provided to older adults
2. changes over time in the individual's sensory processes
3. economic growth and decline, which influence spending power
4. improvement in prenatal care through advances in medicine

Time of measurement effect

People in the year 1999 were concerned about the “Y2K” effect and how it would affect their businesses. If scores on measures of anxiety among 40-year-old business executives taken at this time were unusually high, this might indicate which type of effect in developmental research?

1. attrition
2. time of measurement
3. age
4. time of life

Single-factor research designs

Longitudinal study

In a longitudinal study, a researcher uses which type of approach?

1. follows a group composed of the same people over time
2. asks people to complete an intensive interview over a two-hour period
3. compares people born at different times on one test occasion
4. asks people to try to recall the way they were when younger

Optional- if showing 56 Up

In the film “56 Up,” shown in class, the character Neil Hughes talked primarily about the struggles he’s faced over his life in the area of:

1. racial discrimination.
2. psychological health.
3. relationships with children.
4. weight control.

Prospective study

In a prospective study of the effect of downsizing on workers in their 50s, what approach would the researcher take?

1. randomly select downsized workers and compare them with people currently employed
2. ask downsized workers to complete questionnaires as if they were still employed by the same company
3. survey a large sample of workers in their early 50s and follow them yearly so that those who lose their jobs can be compared to those who remain employed
4. ask downsized workers to keep daily records of their activities, emotions, and attitudes toward their former employers

Selective attrition

In selective attrition, a problem in longitudinal research, the ______ are most likely to drop out of the study prior to the end?

1. most interested
2. least healthy
3. most capable
4. least attractive

Selective attrition

In selective attrition, the researcher’s concern is that the _______ participants tend to drop out before the study is completed.

1. younger
2. less able
3. wealthier
4. happiest

Selective attrition

When a sample of people are followed over time, the less able ones tend to drop out of the sample, a problem known as:

1. quasi-experimentation.
2. cohort reduction
3. selective attrition.
4. logistic regression.

Longitudinal research - cohort

One of the fundamental problems involved in the typical longitudinal (i.e. single-factor) study is that all participants come from the same:

1. social class.
2. nationality
3. cohort.
4. educational level.

Cross-sectional research

In a cross-sectional design comparing adults of different ages, researchers need to attempt to control for which possible factor that can complicate interpretation of the findings?

1. cohort differences
2. time of measurement differences
3. outdating of study methods
4. difficulty in tracking participants

Cross-sectional research

Cross-sectional studies of development compare people of different _________ at one point in time.

1. ethnicities
2. education levels
3. years of birth
4. historical periods

Sequential designs

Cohort-sequential design

In addition to age, a cohort-sequential study manipulates:

1. time of measurement
2. cohort
3. condition
4. time of testing

Time-sequential design

In addition to age, a time-sequential study manipulates:

1. time of measurement
2. cohort
3. condition
4. time of birth

Cross-sequential design

In addition to cohort, a cross-sequential study manipulates:

1. age
2. time of measurement
3. condition
4. time of birth

Most Efficient Design

In the ____________ Design, researchers attempt to investigate which three factors related to development?

1. Quasi-Experimental Design
2. Most Efficient Design
3. Prospective Follow-Up
4. Validity Evaluation

Correlational designs

Correlational design

In a correlational design, the strength of the relationship between two variables is indicated by:

1. A verbal description of what was observed.
2. The percent of the sample with high scores on both variables.
3. A mathematical score ranging from +1.0 to -1.0.
4. The value of the dependent variable.

22. Latent variable

In a study examining the relationship between health and well-being, a researcher chooses to use three indexes that will become the measure of health rather than ask participants to rate their own health. In this case, health is considered a(n) ___________ variable.

1. observed
2. manipulated
3. latent
4. sequential

23. Latent variable

A latent variable in structural equation modeling is one that the researcher:

1. observes.
2. manipulates.
3. disguises.
4. constructs.

Latent variable

As shown in the example of a study linking health to well-being, the three variables of blood pressure, weight, and mobility limitations formed what is known as a(n) __________ variable.

1. multiple
2. latent
3. average
4. predictor

Correlation interpretation

In a study comparing adults from ages 25 through 85 on a test of vocabulary, researchers observe a positive correlation between age and vocabulary test scores. What can the researchers conclude from this finding?

1. Age caused better vocabulary scores.
2. The older the adult, the better the vocabulary scores.
3. The younger the adult, the better the vocabulary scores.
4. People with better vocabulary scores were healthier.

Correlation interpretation

Using the scores from an adult sample ranging from 20 to 71, a researcher reported that there was a correlation of +.63 between a measure of “Meaning in Life” and another of “Self-Acceptance.” What can the researcher conclude about the relationship between these two sets of scores?

1. Higher self-acceptance caused higher meaning in life.
2. Lower self-acceptance is related to higher meaning in life.
3. Higher meaning in life caused higher sense of self-acceptance.
4. The two variables are related but one did not cause the other.

Path analysis

Many researchers in the field of aging are turning to multivariate techniques, such as ____________, to overcome the problems of correlational methods.

1. archival methods
2. path analyses
3. case reports
4. sequential designs

Path analysis

The method researchers use when they wish to study the impact of a set of variables on each other to see if they can be explained by a single model is known as:

1. mediation
2. path analysis
3. simple correlation
4. meta-analysis

Logistic regression

Researchers who try to understand the causes of disease may group participants into those who develop the disease and those who don’t, and test statistical models to understand which factors place an individual at risk. This approach of predicting disease outcomes from a set of risk factors uses which statistical approach?

1. latent variable analysis
2. simple correlation
3. logistic regression
4. focus groups

Logistic regression

In the Health and Retirement Study, researchers used logistic regression to analyze data from a large sample of older adults as they were about to retire. Such a method would permit them to answer which research question:

1. Is a relationship between age and income?
2. Are blue collar workers more satisfied in their jobs?
3. Do men and women feel the same about retirement?
4. Who is most likely to choose early retirement?

Multiple regression

What is the primary objective of research using multiple regression?

1. comparing the mean scores of different groups
2. looking at the relationship between two variables
3. predicting an outcome from a set of variables
4. examining the underlying structure of a particular variable

Multiple regression

Difficulty: Hard

Which type of statistical analysis permits researchers to evaluate the extent to which various measures predict an outcome variable?

1. prevalence study
2. correlational study
3. multiple regression
4. meta-analysis

Mediation

In a study testing the relationship between age and life satisfaction, a researcher introduces a third possible variable of meaning in life that might intervene by influencing the extent to which age alone predicts life satisfaction. In this case, the researcher would be testing meaning in life as having what type of effect?

1. moderation
2. structural
3. qualitative
4. mediation

Types of research methods

Prevalence

An epidemiological researcher is interested in understanding the origins of chronic disease in early adulthood. To establish a baseline of the percentage of people who ever had the disease up through adolescence, the researcher would use which type of statistic?

1. incidence
2. observational
3. correlational
4. prevalence

Prevalence

A researcher discovers that the lifetime prevalence of diabetes in middle adulthood is approximately 13%. This means that for people in this age group:

1. people with diabetes are more likely to die at a younger age.
2. the risk of new cases of diabetes in this age group is 13%.
3. 13% of the population already has a diagnosis of diabetes.
4. diabetes is less of a problem for middle-aged than for young adults.

Incidence

Epidemiological research uses data on the emergence of new cases of a disease to estimate which statistic?

1. regression
2. incidence
3. correlation
4. prevalence

Incidence

The CDC reported in 2019 that 380 new cases of vaping-related lung disease were reported in 38 states of the U.S. This number represents what type of epidemiological statistic?

1. incidence
2. significance
3. prevalence
4. remission

Focus group

A political consultant is working for a candidate to determine what will influence the candidate’s popularity in the voting district. Rather than using a poll, the consultant brings together sets of voters in groups of 10-12 to hear their concerns. This type of method is known as a(n):

1. case study
2. focus group
3. archival analysis
4. observational study

Case report

Difficulty: Hard

In the case report method, researchers on aging gather data by:

1. examining large amounts of information from previously established sources.
2. studying intensively one or a small number of individuals using many measures.
3. giving questionnaires to large, nationally representative samples of adults.
4. asking several people to provide candid reactions to a given problem or situation.

Laboratory research

Which type of research method in the study of aging would be most appropriate for understanding the impact of speed of presenting stimuli on the accuracy of remembering that information?

1. laboratory
2. focus group
3. diary study
4. observational

Observational study

A researcher wishes to study the impact of sidewalk width on how often older adults accidently bump into other people, comparing rush hours with quieter times of day. The best method for completing this study would be:

1. laboratory
2. case study
3. observational
4. archival

Meta-analysis

In order to evaluate the effects of a particular type of therapy on depression in older adults, a researcher examines the results of 155 studies that included 19,000 individuals. The examination through statistical methods of such a large number of investigations is known as what type of method?

1. qualitative analysis
2. focus group
3. meta-analysis
4. experimental manipulation

Meta-analysis

In a meta-analysis, researchers analyze data from which sources?

1. responses by individuals to questionnaires
2. health records of individual participants
3. multiple studies on the same topic
4. careful inspection of the data from one investigation

Diary study

A diary study would be particularly useful for researchers who wish to:

1. observe their participants in controlled conditions.
2. follow the daily exercise habits of their participants.
3. study a few people very intensely.
4. ask people to answer a questionnaire about current events.

Survey

In which type of research method would the data produced involve having participants complete a set of 1-7 point rating scales in which they respond to items about themselves?

1. case study
2. survey
3. focus group
4. meta-analysis

Measurement issues in adult development and aging

Test-retest

Measures with high ___________ reliability provide similar results over a brief period of time among the same participants.

1. criterion
2. construct
3. observational
4. test-retest

Validity

In a study on job performance and age, an occupational psychologist asks workers at a large international company to complete online measures in which employees and their supervisors both complete their own versions of tests of the employee’s performance on a set of rating scales. The correspondence between the two measures could be considered an indication of the test’s:

1. validity
2. reliability
3. consistency
4. utility

Ethical issues in research

Informed consent

An outpatient medical office switches to a new system of scheduling appointments that involves using a smartphone app. The healthcare professionals who run the office decide to submit a research paper on the new system to a national conference for which they will use patient satisfaction data. To ensure that the study is conducted ethically, what step must the researchers follow?

1. Advertise the study in a patient newsletter.
2. Give patients reduced rates for participating.
3. Require patients to use the app before taking phone calls.
4. Ask patients to provide their informed consent.

Debriefing

The ethical requirement of debriefing means that researchers are required to take what step with their participants?

1. Provide them with general feedback about the nature of the study.
2. Reveal any problems that their scores revealed about them.
3. Ask them to sign a form at the beginning of the study.
4. Obtain their permission before publishing the study’s results.

Anonymity

Which ethical requirement states that researchers should not associate names of participants with their responses?

1. debriefing
2. anonymity
3. confidentiality
4. informed consent

Short answer questions

Variables in developmental research

1. Compare an experimental with a quasi-experimental design. Which type of design is used in studies on aging?
2. Define age, cohort, and time of measurement. What is each variable intended to represent?
3. Why are age, cohort, and time of measurement dependent on each other? Why is this a problem in research on adult development and aging?

Single-factor research designs

1. Define a longitudinal and a cross-sectional design. Compare two advantages and two disadvantages of each.
2. What solutions have researchers developed to overcome the problems of descriptive research designs?

Sequential research designs

1. What is the purpose of sequential research?
2. What are the three sequential designs in the Most Efficient Design?

Correlational approaches in research on aging

1. What is the reason that researchers do not allow cause-and-effect conclusions to be drawn from correlational studies?
2. List and define two multivariate correlational approaches. How are these used to overcome the problem that "correlation does not equal causality"?

Types of research

1. Provide examples of each of the following methods:
   1. Case study
   2. Observational study
   3. Focus group
2. You are hired to be a consultant in a nursing home to study socialization patterns of residents. Which research method would give you the most useful data? Why?

Measurement issues in adult development and aging

1. What are the two forms of reliability? Why are each important?
2. List the four forms of validity and provide an example of how each would be measured.
3. Describe why it is important for researchers to establish the reliability and validity of measures for each age group in a cross-sectional study.

Ethical issues in research

1. What are the four components of ethical standards in research?
2. Why were ethical standards developed and why are they important?

Essay questions

1. Design a study on a topic of interest to you using longitudinal and cross-sectional methods. Describe the potential problems you would have with each method and how you would overcome these problems.
2. Describe how you would apply the sequential designs to a study you develop. Analyze how the pattern of results you obtain would indicate the effects of age, cohort, and time of measurement.
3. Discuss the problems involved in simple correlational designs and explain the advantages to using multivariate designs instead. Think of a research example to illustrate these points.
4. Analyze the issue of psychometrics in research on aging. Why is it important for researchers to ascertain that measures are equally applicable to each age group in a study? Do the same considerations apply in longitudinal research designs?
5. Describe how you would go about ensuring the rights of participants in a study on memory and aging. Design a consent form, a debriefing form, and a set of guidelines to use that will ensure protection of the confidentiality of data.