Careers and Research in Health Psychology



Careers and Research in Health Psychology

September 16, 1999

The Biopsychosocial Perspective

Biomedicine vs. the BPS Perspective

Treat a disease and you may win or you may lose; treat the whole person and you win every time, no matter what the outcome… Patch Adams

Health Psychology’s Perspectives

The BPS Perspective

The Life-Course Perspective

Cohort effects

The Sociocultural perspective

The Gender perspective

Career Issues: Roles and Training

3 roles

Teachers

Research scientists

Applied clinicians

Training

Ph.D. or Psy.D. programs

Career Issues: Where Do Health Psychologists Work?

Conducting/Evaluating Health Psychology Research

Scientific and Not-so-Scientific Reasoning

Faulty intuition

Overconfidence

Belief bias

Confirmation bias

Hindsight bias

Scientific Reasoning

Skepticism

Empiricism

Objectivity

Openness to alternative explanations

Self-correcting

replication

Humility

Principle of Parsimony

Conducting Research in Health Psychology

The Role of Theory

A set of related assumptions from which testable hypotheses can be drawn

Theories organize data and guide research

Hypotheses create operational definitions of independent variables and dependent variables

The Research Loop

Descriptive Research

Observational Studies

Naturalistic observation

Field studies vs. lab studies

Case studies

Self-Report Studies

Surveys and questionnaires

Interviews

Correlation v. Causation

Correlation coefficient

-1.0 < r < + 1.0

Strength (absolute value)

Direction (sign: positive/negative)

Positive Correlation

Negative Correlation (r = -.70)

No Correlation (r = 0)

Correlation and Causation

Experimental Research

cause-effect relationships between independent variables and dependent variables

Between Groups Design

At least 2 groups tested on the effects of at least 2 levels of at least one independent variable

Experimental group (e.g. drug)

Control group (e.g. placebo)

Dependent variable (hypertension)

Within Subjects (Groups) Design

Each subject serves as his/her own control

Quasi-Experimental Research

Ex post facto design

Compares two groups that differ on the independent variable under study at the start of the study

Independent variable is not manipulated

E.g. sedentary vs. active individuals and CHD

Other ex post facto variables: age, gender, ethnicity, socioeconomic status

Lifespan Change Methods

Cross-sectional study

Cohort effects

Longitudinal study

Cross-sequential study

Epidemiological Studies

Observational in nature

Cannot establish cause and effect

Determining associations between specific risk factors and disease

risk factors—characteristic or condition that occurs more often in people with a disease than in people free from the disease

Test whether hypotheses drawn from other studies are consistent with epidemiological data (e.g. identifying

Provide a basis for evaluating different preventive procedures

Areas of Epidemiological Study

Observational Methods

Retrospective study (retro-backward)

Also called case-control studies. Compare background risk factors in cases of a disease with controls

Prospective study

Also called cohort studies. Follow healthy people forward in time.

Framingham Heart Study

Nurses Health Study

Prospective Study of 250 medical students/MDs followed 25 years)

Epidemiological Methods

Natural Experiments

Similar to ex post facto design. Two groups of people naturally divided themselves into those exposed to a pathogen and those not exposed.

Clinical Trials

Similar to experiment. Randomly assigns subjects to two or more treatment groups

Example of Epidemiological Research

The Alameda County Study

Large community trial begun in 1965 to identify certain health practices that related to mortality and morbidity

People who practiced 6 or 7 basic health-related behaviors had lower mortality and morbidity rates than those who practiced zero to 3 of these behaviors

getting enough sleep; eating breakfast; rarely eating between meals; using alcohol in moderation or not at all; not smoking; exercising regularly; maintaining ideal weight

Validating Research

Validity & reliability of observations/measures

Wording of survey questions (framing effects)

NYT study: “oppose abortion” v. favor “protecting the rights of the unborn child”

Avoiding Sampling Error

Samples, populations

Representative v. unrepresentative (biased samples)

Random samples/stratified samples

Sample size

Validating Research

Observer-expectancy effect

Subject-expectancy

Blind and double-blind controls

Standardizing (“norming”) tests

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download