EXS 407: Research and experimental design

What is the definition of research?

The systematic investigation into and study of materials and sources to establish facts and reach conclusions.

What is the key to formulating a research question?

Carefully formulating a research question or falsifiable hypothesis.

What is observational research?

Research where the investigator measures relationships between events or conditions without manipulation, also known as descriptive research.

What is experimental research?

Research where the investigator directly manipulates conditions to measure the response.

What are independent variables (IV)?

Variables controlled by the investigator, also known as the uncontrolled 'cause' or treatment.

What are intervening or confounding variables?

Variables that influence the dependent variable but are not controlled by the investigator, potentially leading to erroneous conclusions.

What is the goal of sampling in research?

To extract a sample that is representative of the population.

What is a random sample?

A sampling method where each member of the population has an equal chance of being selected.

What is a stratified sample?

A sampling method that ensures representation of subgroups within the population of interest.

What is a convenience sample?

A sampling method where members are selected based on ease and proximity, often leading to bias.

What is bias in sampling?

When aspects of the sample make it unrepresentative of the population.

-larger sample=less bias

What is a systemic sample?

A sampling method where members are selected at regular intervals from a randomly-ordered list.

ex: every 6th person

What is a cluster sample?

A sampling method where populations are divided into clusters, and members are randomly selected from a cluster.

-common for studies of large populations, ex: recruiting from Omaha

What are pre-experimental designs?

-exploratory

- provide weak evidence of causality

-lacking control groups or random assignment.

ex: case study, pretest-posttest, static group comparison, nonequivalent groups

quasi-experimental designs

-moderate evidence of causality

-no random assignment (not possible/ethical)

ex: interrupted time series, natural experiment

What are true experimental designs?

Research designs that involve random assignment of participants to treatment or control groups, providing strong evidence of causality.

ex: independent groups, matched groups, RCT, repeated measures, factorial, pretest-posttest, soloman four-group

What is a case study in research?

An observational method where a single group is exposed to an intervention and the outcome is measured.

-can be lots of confounding variables

What is a pretest-posttest design? (one group)

A design where a single group is measured before and after an intervention

-may be confounding variables

What is a static group comparison? (2 groups)

A design where two groups are compared, but only one receives the treatment, and individuals are not randomly assigned.

-preexisting differences may effect result

What is an interrupted time series design?

A design that takes multiple measurements before and after an intervention to examine changes in trends.

-no control over confounding variables

What is a natural experiment?

-observes the effects of natural occurrences or changes

-limited/no control over intervention

-high external validity because its natural

What is a randomized controlled trial (RCT)?

-participants are randomly assigned to treatment or control groups, often using blinding to reduce bias.

-single blind: participant unaware of group

-double blind: participant and researcher unaware of group

-gold standard for clinical research

What is the Solomon four-group design?

A design that separates treatment and control groups, which may or may not be pretested

-improved internal validity

-controls for carryover effects from the pretest

-requires a larger sample size and random assignment

What is factorial design in research?

-examines the effects of multiple IVs on a single dependent variable, allowing for main effects and interactions.

-larger sample size needed

repeated measures design

Participants complete all conditions, serving as their own controls (random/counterbalanced order of treatments to minimize carryover effects) which requires smaller sample sizes.

-more sensitive to effects of IV

independent groups design

-random assignment to study group (minimizes effects of individual differences)

-individual differences can still influence the result

matched groups design

-participants are matched on key attributes (age, weight, etc), then randomly assigned to groups (minimizes effect of individual differences further than ind. groups)

-useful when specific attributes are expected to interact with the IV

pretest-posttest (true experimental)

-control group/placebo tells us what changes/effects of IV, controls for confounders

-compare pretest and posttest between placebo and treatment group