exam 3 study pt 1

Observational & Single-Subject Research Designs ppt

what is the purpose of an observational study and when would you use it

Observe what happens naturally without giving a treatment.

Researchers do not manipulate anything—they only collect information.

Use when:

• You want to identify patterns or relationships.

• Giving a treatment would be unethical or impossible.

• You're generating hypotheses for future studies.

Key point: Observational studies cannot prove cause and effect because no intervention is given.

what is a prospective study, what threatens it

Present → Future

Researchers begin now and follow people over time.

threat» attrition

(more control clear sequence)

what is attrition

the gradual, steady reduction of something's size, numbers, or strength.

what is retrospective study and what threatens it

study that mines existing records/databases

threat»missing/inconsistent data

(cheaper, faster, larger N)

Define Cross sectional vs longitudinal

cross sectional» •data from different sources collected at same time; selection bias= biggest risk

longitudinal» •data from same source collected over period of time; watch for attrition & testing effects

correlational study def

•strength/direction of association ®

Correlation does not equal causation

difference between the case control and cohort observational study designs

Case-control - studies start with the outcome (disease) and look backward for exposures,

cohort studies - start with the exposure and follow participants to see who develops the outcome.

what are common biases that occur

• Selection bias: Participants aren't representative.

• Recall bias: Participants remember past events inaccurately.

• Interviewer bias: Interviewer's behavior influences responses.

• Confounding: Another variable affects both exposure and outcome.

what fixes can someone do to reduce bias

• Matching → balances confounders between groups.

• Blinding chart reviewers → reduces reviewer bias.

• Statistical adjustment → controls confounding.

• Triangulation → uses multiple methods/data sources to confirm findings.

what are the single subject design core elements

• Repeated measures = Measure many times, not once.

• A = Baseline (no treatment).

• B = Intervention (treatment).

• Graph everything = Plot every data point and look for changes at the phase line.

describe an A-B single study design

•Quick clinical feedback but weak internal validity : no control comparison

•Impossible for this design to claim causation

how can someone strengthen an A-B SSD

•repeat phases after return to baseline

•Effects more than one subject or more than one target outcome

• by Comparing 2+ interventions

Describe an A-B-A ssd

Shows how/if behavior reverses when treatment is removed. ( does this by adding a second baseline

Describe an A-B-A-B ssd

GOLD STANDARD for reversible behaviors

•two chances to see effect

describe an alternating treatment ssd

Compares two interventions to determine which will be more effective

» also used for control and placebo treatments

describe an A-B-C-A ssd

compare two interventions sequentially after baseline

describe direct replication

• Same design

• Same type of participant

• About 3 successful replications strengthen confidence.

describe systematic replication

Change one factor:

• Different setting

• Different therapist

• Different age group

Shows results generalize.

external validity

transferability>

• "thick" descriptions so that future researchers or practitioners can determine if the findings are applicable to them own specific settings or contexts.