Research Methods Midterm

Historical Context of research

Empirical approach (direct observation)

Social Cultural context of research

Influences researchers choice of topics, society's acceptance of findings, and the locations in which research takes place.

Moral context of research

Demands researchers maintain highest standards of ethical behavior

Multi-method approach to research

Search for an answer using various research methodology and measures of behavior

independent variable

The experimental factor that is manipulated

Constructs

Concepts or ideas (Intelligence, depression, aggression, and memory)

Operational Definition

Explains a concept solely in terms of observational procedures used to produce and measure it

Reporting

What separates what you have observed from what you infer

Reliability=

Consistency

Reliability:

Stability or consistency of measurement.

Validity:

how well a test actually measures what it is supposed to measure — and whether we can make accurate decisions from it.

Validity

Accuracy

Goals of the scientific method

1. Description (seek to describe events/relationship)

2. Prediction (hypothesis)

3. Explanation (understanding cause)

4. Application (Apply knowledge and research methods)

What is the IRB

• Institutional Review Board

  • It reviews research proposals to ensure ethical considerations are met.

Minimal Risk in research is-

When procedures or activities in the study are similar to those experienced in everyday life

Informed Consent

Person explicitly expressed willingness to participate in a project based on clear understanding of research

Privacy

Rights of the individual to decide how info about them is communicated to others

Steps for ethical compliance

1) Review face of the proposed research situation
2) Identify the relevant ethical issues
3) Consider multiple viewpoints
4) Consider alternative methods or procedures and their consequences

Why should we care about research?

Foundation of psychology, and relevant to everyday life

What do procedures do in research?

Create information (researchers, academics, etc)

What do consumers do in research

"interrogate" information (therapists, teachers, etc)

Names of the Four Scientific Cycles

1. Theory data cycle

2. Basic-Applied research cycle

3. Peer-review cycle

4. Journal to journalism cycle

The Four Scientific Cycles:

• Theory-Data cycle

Scientists collect data to test, change, or update theories

The Four Scientific Cycles:

• Theory-Data cycle

  • Theory-

A statement that describes general principles about how variables relate to one another.

The Four Scientific Cycles:

• Theory-Data cycle

  • Empiricism-

Collect data to figure out or challenge theory

The Four Scientific Cycles:

• Theory-Data cycle

  • What makes a good theory

1. Supported by data

2. Falsifiable

3. Parsimonious ( all things being equal)

The Four Scientific Cycles:

• Basic-Applied research cycle

  • Basic research:

  • Applied research:

• Basic research: goal is simply to enhance general knowledge

• Applied research: done with practical problem in mind; research will be directly applied

The Four Scientific Cycles:

• Peer-review cycle:

Submission -> review -> feedback -> revisions -> review

The Four Scientific Cycles:

• Journal to journalism cycle

Popular press picks up a topic in the field

Research vs. Experience-

• Experience has NO comparison group

  • Ex: punching a punching bag to release anger vs. sitting quietly to release anger

The good story/it makes sense-

Cognitive bias

• Punching a punching bag just makes sense

The present/present bias-

Cognitive bias

• Can't remember times when you didn't punch a punching bag

Pop-up princple/availability heuristic-

Cognitive bias

• Information that is more salient is what we're more likely to make judgements about

• Ex: fearing plane crashes more than car accidents because plane crashes receive more intense, memorable media coverage, despite being statistically less frequent. 

Cherry picking the evidence/confirmation bias

Cognitive bias

• Seeing what we want to see

Asking biased questions-

Cognitive bias

Being overconfident in research-

Cognitive bias

• Not a good indicator of accuracy

Confounds:

• Alternative explanations

• 3rd variable

Tuskegee Syphilis Study:

• Government study from 1932-1972 which investigated effects of untreated syphilis on African American males.

• They were told they were being treated, when they weren't.

Milgram obedience study:

• Ethical considerations:

  • stressful to participants

  • lasting effects of study

Beneficence (ethics):

• Cost-benefit analysis

  • Does it do more harm than good

Belmont report:

• Respect for persons: informed consent

• Respect for Beneficence: cost benefit analysis for participants and society

• Respect for Justice: How are participants selected

APA ethical princples (5):

1) Beneficence and non-maleficence
2) Fidelity and responsibility
3) Integrity
4) Justice
5)Respect for people's rights and dignity

Deception:

• Through omission:

• Through commission:

• Through omission: Leaving out info

• Through commission: lying

Frequency Claim

A statement about how often someone does something

Association claims-

• Positive:

• Negative:

• Curvilinear:

• Zero:

• Positive: as one increases the other increases, vice versa

• Negative: As one increases, the other decreases

• Curvilinear: inverted u; positive up to a certain point then negative

• Zero: no relatinoship

Causal Claim:

One variable causes another

Dependent Variable:

What's being measured

Conceptual definition:

Researcher's definition of a variable at an abstract level

Construct validity:

How well were all the variables are measured and manipulated

External validity:

To whom or what can you generalize theclaim

Statistical validity:

How well does your data support the conclusion

Internal validity:

Are there alternative explanations for the outcomes

Co-variance

Show the statistical significance between both variables (As one changes, the other does too)

Temporal Precendence:

A comes first in Time before B

Randomized experiment:

• Manipulation of IV

• Random assignment

• Measurement of DV

Quasi experimental:

• Manipulation of IV

• Nonrandom assignment

• Measurement of assumed DV

Non experimental:

• Measurement of assumed IV

• Nonrandom assignment

• Measurement of assumed DV

Are survey and poll inter changeable?

Yes

Survey and poll:

• Do you need a large sample or a high response rate?

Not necessarily; representativeness is more important.

Survey and poll:

• Pre-testing:

Piloting a survey before the actual experiment; detects possible problems beforehand

Survey and poll:

• Interviewing style:

Rigid vs. conversational; opportunities to clarify meaning can increase validity

Survey and poll:

• Open-ended questions

Rich information, but requires coding

Survey and poll:

• Closed/forced choice format

• Respondent picks from options

• Gather presence or absence of constructs listed

Survey and poll:

• Optimizing

The way we want people to process surveys. Search memory for relevant information & integrate into judgement; translate judgement into response.

Survey and poll:

• Satisficing:

• Revised response strategy that requires less effort; mental shortcuts; people are fatigued/questions are too advanced/etc

• a respondent's tendency to provide low-effort or "good enough" answers, rather than putting in the necessary mental work to provide highly accurate, optimal responses.

Survey and poll:

• Conversational conventions:

Norms and expectations about everyday conversations influence the interpretation of question and response meanings (EX: info presented at the beginning is unimportant, so people skim it)

Survey and poll:

• Response alternatives in questions can...

Clarify intended meaning of question or focus, Remind correspondents of material they may not otherwise consider;

Survey and poll:

• Other considerations for developing questions for participants

• Context matters (give info they THINK the researchers want)

• Order effects

• Adjacent questions/subsequent judgment

• Social desirability

• Acquiescence

Survey and poll:

• Primacy & Recency effects-

the tendency to show greater memory for information that comes first or last in a sequence

Survey and poll:

• Adjacent questions/ subsequent judgement-

Content of earlier questions influence interpretation of later questions

Survey and poll:

• Social desirability:

Desire to present oneself in a positive manner

Survey and poll:

• Acquiescence

Tendency to endorse any assertion made in a question, regardless of its content

Writing Well-Worded Questions:

• Leading question-

• May prime or bias the respondent;

  • "Do you think that relations between Blacks and Whites...

  • Will always be a problem or a solutions will eventually be worked out (negative)

• VS

• Are as good as they're going to get or will they eventually get better (positive)

Writing Well-Worded Questions:

• Double barreld questions-

Asked two questions in one

Writing Well-Worded Questions:

• Double negative

Difficult to interprets; "I wouldn't never do that to someone"

Writing Well-Worded Questions:

• Question order matters?

Can provide meaning; can prime

Response Set Types

• Yeah saying-

Saying 'yes' or 'strongly agree' to everything

Response Set Types

• Nay saying:

Saying 'no' or 'strongly disagree' to everything

Response Set Types

• Fencing sitting

Answering in the middle

Correlational research-

Provides basis for making predictions

How are surveys used?

Assess people's thoughts, opinions, and feelings

Response rate bias

Threat to the representativeness of a sample that occurs when some participants selected to respond to a survey systematically fail to complete the survey (e.g., due to failure to complete a lengthy questionnaire or to comply with a request to participate in a phone survey).

Sampling-

Careful selection of participants that allow researchers to generalize findings from the sample to the population

Representativeness-

The ability to generalize from a sample to a population

Selection bias-

When procedures used to select the sample result in over-representation of some segment of the population

probability sampling-

Selecting people randomly from a list

Probability sampling- Selecting people randomly from a list

• Sampling frame:

a list of all of the members of a population

Non-probability sampling-

Non-randomly selecting people

Probability sampling

• Simple random sampling-

Each element of the population has an equal chance of being included in the sample

Probability sampling

• Cluster sampling

randomly select clusters of participants from population, then include every element in that cluster

Probability sampling

• Cluster sampling

  • Multistage sampling-

• 2 random samples are taken from population –

  • cluster from population, then sample from cluster

Probability sampling

• Stratified random sampling-

Divide population into subpopulations then randomly select particiapants

Probability sampling

• Stratified random sampling- Divide population into subpopulations then randomly select particiapants

  • Strata:

Subpopulations of interest (young, middle age, old)

Probability sampling

• Stratified random sampling- Divide population into subpopulations then randomly select participants

  • Oversampling:

One or more groups are intentionally overrepresented

Probability sampling

• Stratified random sampling- Divide population into subpopulations then randomly select participants

  • Why would a researcher use oversampling?

to improve the representation of minority classes in imbalanced data

Probability sampling

• Systematic sampling

choosing every nth person in a population

Are larger samples better?

Not necessarily; not useful if not representative.

Generalizability:

How well does the relationship you found in your sample represent the relationship in the population?

Non-probability Sampling include:

accidental sampling; quota sampling; purposive samples; snowball sampling

Non-probability sampling-

• Convenience/accidental samples

Data is collected from the cases at hand until the sample reaches a designated size