PSYC 101 Ch. 2 - Research Methods

Intuitive thinking

snap judgements, quick & reflexive

Analytical thinking

slow & reflective, e.g. learning a new skill

Heuristic

Mental tendency to favor intuitive thinking

Replication crisis

a systematic problem in which a large proportion of studies are not reproducible

Naturalistic observation

watching behavior in real world situations with no manipulation

high external validity

low internal validity

Case studies

examining 1 person or a small group in depth over an extended period of time

existence proof

anecdotal

low internal validity

Self-report

e.g. questionnaires, surveys, interviews

easy & cheap

qualitative data w/ high accuracy of true

hard to interpret

risk response set, bias & narcissism

Correlational design

examine extent of correlation (positive, negative or zero)

can be used to predict behavior

low internal validity

Experimental design

randomly assigned participants

highly controlled conditions

manipulate independent variable

high internal validity

sometimes low external validity

Quasi experimental design

Study causal effect but lacks random assignment of subjects to conditions

No control

Subjects selected based on their pre-existing characteristics instead of being assigned to them by researcher

Still an 'intervention'

Cannot deduce causal relationship

existence proof

demonstrate that a psychological phenomena occurs

response set

tendency of participants to distort answers to surveys

experimenter expectancy effect

Hypothesis leads to unintentional bias in outcome of study

demand characteristics

subjects guess hypothesis of study through cues resulting in bias

informed consent

subjects know what is involved in a study before being asked to participate:

- Purpose of study

- Procedures used

- Risks & Benefits

- Rights to refuse & terminate participation

- Compensation

- Discussion of protection of confidentiality

- Contact information

debriefing

subjects told point of a study at the end

Tri-Council Policy Statement (TCPS) Ethical Codes

1. Concern for Welfare(Beneficence) -> Protect against harm or discomfort

2. Respect for Persons(Autonomy) -> Informed consent

3. Justice -> Subject selection

Descriptive statistics

e.g. distribution curves

DESCRIBES data

Summarize mass data points

Understanding & interpreting data

Provide visual display for appropriate calculations

Use of descriptive statistics in experimental

Compare mean of two grps

Use of descriptive stats in correlation

Significance of r value

Mean

Average of all data

Affected by outliers

Properties good for statistical analysis

Median

Value splitting all data down the middle

Used when outliers present to show what is "in the middle"

E.g. household income

Mode

Most frequent value

Variability

How spread out data is

Range (max-min)

Standard deviation

Standard deviation

Uses all data

Standard diff between scores e.g. 15IQ

Good for stat calculations

Larger S.d. = less equal across

normal distribution

A function that represents the distribution of variables as a symmetrical bell-shaped graph.

Best used with s.d.

68% in 1s.d. From mean

95% in 2s.d. From mean

99.7% in 3s.d. From mean

Inferential statistics

numerical data that allow one to generalize- to infer from sample data the probability of something being true of a population

e.g. hypothesis testing

central tendency

where group tends to cluster

base rate fallacy

tendency to ignore general statistics to focus on specific information when making judgements

Evaluating research

1. Method

2. Peer review

3. Source

4. Extent of levelling & sharpening

5. 'Balanced' coverage

sharpening

exaggerating main point of study

levelling

minimizing less central data

Pseudosymmetry

tendency in media reports to try to present two opposing sides of an issue as though both sides were equally valid

Why people believe in ESP

Disproved with replication

People underestimate the likelihood of coincidences and cold guesses

Animal research debate

Clear benefits to understanding human physiology, bio and psych

Poor conditions of keeping and feeding animals

Large no. killed each year

No good alternative

internal validity

Causal relationship

External validity

Generalizability of findings to other contexts.

W estern

E ducated

I ndustrialized

R ich

D emocratic

Ecological validity

Specific part of external validity - how applicable findings are to real world settings

Ways to operationalize a variable

Situation, response participant

Situation variable

characteristics of a situation or environment that can be measured/manipulated

Response variable

Response(e.g. reaction time, Q&A)/behavior. Only measured

Participant variable

Characteristics that individuals bring with them. Only measured

Benefits of Research Participation

- Material

- Education

- Treatment

- New skills

- Helping society

- Consequences of not participating

Risks of Research Participation

- Physical, emotional or psychological harm

- Vulnerability

- Social Risk

Inclusion/Exclusion of subjects

- Scientifically justifiable

- Group must benefit

- Group must not bear all burden of research