PSY 240 EXAM 4

How are surveys/interviews different from other forms of research?

Surveys/interviews collect self-reported data emphasizing opinions and behaviors rather than observed outcomes.

What can be studied with surveys?

Attitudes, beliefs, behaviors, demographics, preferences, and psychological constructs.

Strengths and weaknesses of mail surveys

Strengths: inexpensive, broad reach. Weaknesses: low response rate, delayed responses.

Strengths and weaknesses of internet surveys

Strengths: fast, cost-effective, wide access. Weaknesses: sampling bias, tech barriers.

Strengths and weaknesses of group surveys

Strengths: efficient, consistent environment. Weaknesses: peer influence, limited generalizability.

Strengths and weaknesses of phone interviews

Strengths: clarification possible, personal touch. Weaknesses: costly, time-consuming.

Strengths and weaknesses of personal interviews

Strengths: rich data, flexible. Weaknesses: expensive, time-intensive.

How to counteract weaknesses in survey methods?

Use mixed methods, improve clarity, offer incentives, ensure anonymity, use follow-ups.

How should surveys be constructed?

Use clear, concise, neutral questions; pilot test; use logical flow.

Closed vs open-ended questions

Closed: easy to analyze but limited detail. Open-ended: richer data but harder to code.

What are scales in surveys?

Measurement tools (e.g., Likert); should be clear, avoid ambiguity or bias.

Types of questions to avoid

Leading, loaded, double-barreled, ambiguous questions.

Sampling techniques and which are better

Random, stratified, convenience, snowball. Random best for generalizability.

Difference between nomothetic and idiographic research

Nomothetic seeks general laws; idiographic focuses on individual understanding.

Why use idiographic or nomothetic research?

Idiographic for unique cases; nomothetic for broader applications.

Single-subject vs case study

Single-subject: systematic manipulation; Case study: descriptive exploration.

History of single-subject design

Origin in behaviorism, especially B.F. Skinner’s work.

Why use baselines in single-subject research?

Establish comparison; should be stable and consistent.

Different single-subject designs

AB, ABA, ABAB, multiple baseline, changing criterion.

Carryover effects in single-subject designs

Effects from previous condition that influence results, lowering validity.

Challenges in single-subject designs

Generalizability, variability, reactivity; addressed with replication and blinding.

What is a non-reactive measure?

A measure that does not influence participants’ behavior.

What are traces and products?

Traces: physical evidence. Products: artifacts from behavior.

Accretion vs erosion traces

Accretion: build-up (e.g. trash). Erosion: wear (e.g. carpet).

Controlled vs natural traces

Controlled: researcher-managed. Natural: real-world occurrence.

Confounds in non-reactive studies

Ambiguous causality, selective survival of data.

Ethical considerations in trace research

Consent, privacy, potential harm.

Difference between document and record; continuous vs discontinuous

Document: written artifact. Record: systematic log. Continuous: ongoing; discontinuous: periodic.

Processing archival data

Coding, categorizing, quantifying existing records.

Confounds in archival research

Incomplete records, selective reporting, outdated info.

Ethical issues in archival studies

Privacy, consent, sensitivity, misuse.

AB design

A single-subject research design where 'A' is the baseline and 'B' is the treatment phase.

ABA design

A single-subject design that includes a baseline (A), treatment (B), and return to baseline (A) to assess treatment effect.

ABAB design

A single-subject design with two baseline-treatment cycles (A-B-A-B), increasing internal validity.

Multiple baseline design

A design that applies treatment across different behaviors, settings, or subjects at different times to control for confounds.

Changing criterion design

A design where the behavior target changes step-by-step to demonstrate treatment effectiveness.

Likert scale

A psychometric scale commonly used in surveys to measure attitudes by levels of agreement.

Semantic differential scale

A type of scale using bipolar adjectives (e.g., good–bad) to measure connotative meaning.

Double-barreled question

A question that asks about two things at once, reducing clarity and validity.

Leading question

A question that suggests a particular answer, introducing bias.

Loaded question

A question with built-in assumptions that may be controversial or emotionally charged.

Stratified sampling

A sampling method that divides the population into subgroups and samples from each, ensuring representation.

Convenience sampling

A non-random sampling technique that selects individuals who are easiest to reach.

Snowball sampling

A method where existing participants recruit future participants, useful for hard-to-reach populations.

Systematic sampling

Selecting every nth individual from a list after a random start.

Archival research

A method involving analysis of pre-existing data such as records, documents, or databases.

Physical trace measure

A non-reactive method that uses physical evidence of past behavior (e.g., footprints, wear).

questionnaire

a set of questions created to learn about individuals, not meant to be aggregated

sampling bias

occurs when a sample overrepresents some subset of the population and underrepresents other subsets

response rate

calculated by dividing the actual number of responses by the number of potential responses

interviewer bias

the interviewer’s expectations or preferences. reduced in internet surveys

group-administered surveys

given in a setting where it is easy for recipients to complete them, most people will comply. response rate is stronger than for mail surveys

socially desirable responses

reflects what is deemed appropriate by society, but are not necessarily reflections of the respondent’s practices or beliefs

funnel structure

where a survey begins with general, interesting, and easy to answer questions. the questions become more specific and require more thought in the middle of the survey

demographic questions

descriptive questions about the respondent’s social statistics, such as gender, age, income level, etc.

branching

the answers to demographic questions might determine which specific questions the person will see next

filter question

used to determine the next question to ask based on the answer

reliability

the degree to which a measurement tool provides consistent answers

test-retest reliability

measures the degree to which a test generates the same responses upon retesting

alternative-forms reliability

assesses how well two forms of the same test yield comparable results

construct reliability

the degree to which respondents’ replies within the instrument are consistent

split-half reliability

determines if a person’s replies to one half of the items are related to the same person’s replies to the other half of the items

cronbach’s alpha

assesses internal consistency. a statistical technique that compiles the correlations of every item with every other item within the tool

validity

the degree to which a measuring tool measures what it purports to measure

construct validity

the extent to which the concepts thought to be measured within the tool are actually being measured

face validity

the most straightforward. the degree to which a measurement tool appears to be measuring what it is supposed to be measuring

criterion validity

measures how well the results of an instrument correlate with other outcomes or behaviors

pilot study

a small group of people is given the survey, giving the researcher the opportunity to work out any bugs in the survey questions or the data gathering

population

consists of all the members of a given group to which the research is meant to generalize

sample

a subset of the population

sampling frame

a list of all the members of a population

elements

the members of the sample who are chosen from the sampling frame

random selection

all members of the population are equally likely to be chosen as part of the sample

random sample

elements are randomly chosen from a sampling frame

systematic sampling

elements are not chosen randomly; chosen according to some specific plan or strategy

cluster sampling

clusters of potential respondents that represent the population are identified, then all of the people in those clusters are included in the sample

quota sampling

a combination of convenience sampling and stratified sampling

nonprobability sampling

each member of the population is not equally likely to be selected and the outcome could easily be a biased sample

time-series design

several measurements are made before and after the introduction of the independent variable

withdrawal design

a series of baseline measurements is compared to the measurements taken after the introduction of the intervention. the intervention is then withdrawn, and measurements are continued.

ABAB design

a research design involving baseline, intervention, then return to baseline, then again the intervention

ABAC design

introduction of a second intervention. baseline → intervention #1 → baseline → new intervention

ABCB design

can be used to assess the effect of an intervention compared with a placebo condition in the following way. baseline → intervention → placebo → intervention

reversal design

a new and opposite intervention is introduced

alternating-treatments design

a variation of the ABAB design; this design does not require baseline, and two or more treatments are presented to the participant instead of just one. introduction of treatment may be random or systematic

multiple-baselines design

the effect of a treatment ton two or more behaviors is assessed, or the effect of a treatment on a single behavior is assessed across two or more situations

changing-criterion design

avoids withdrawing treatment. used to assess an intervention when the criterion for that intervention is routinely changed

subject bias

what the participant thinks should happen. confound in single-subject designs

demand characteristics

what the participant thinks the research wants to happen. confound in single-subject designs

natural trace measures

trace measures that occur without researcher intervention

controlled trace measures

trace measures that involve researcher intervention

products

these are purposeful creations by individuals

selective survival

refers to the notion that some trace or product evidence may not endure over time

selective deposit

refers to the circumstance in which not all traces are equally representative

record

an account or statement is created for another person to read, view, or hear

document

not created at someone’s request or for someone else to use. considered more personal than records

data reduction

reducing the amount of information to a more usable amount

content analysis

researcher develops a coding system that is used to record data regarding the content of records

intercoder reliability

determines reliability of the coding system