Scheduled maintenance: Wednesday, February 8 from 10PM to 11PM PST
Upgrade to remove ads
Only $35.99/year

Social Work Research

Terms in this set (92)

T/F: NASW code of ethics requires research skills as part of ethical practice. SW have both a responsibility for both evaluating practice and building knowledge.
True
Evidence Based Practice (EBP)
Interventions proven to be effective for clients
T/F: Use a technique because it fits with my belief system, ideological, spiritual, political, etc...
False
T/F: Use a technique because it has scientific support
True
Step 1 of EBP
Forming Questions to answer practice needs. Main problem & what needs to be done?
Formed questions to answer practice needs should
-Target practice decisions
-Consider variations in cl characteristics (culture, religion) -- what intervention, program, or policy has the best effects?
Systematic review
thorough, comprehensive attempt to collect all the information available to answer a research question
Step 2 of EBP
Search for all evidence, not just the kind that supports your view.
Bottom-up searches
Finding additional resources off one source's reference page
Top-down searches
Using all data bases (ebscohost, medline, psyinfo), starting from scratch
Step 3 of EBP
Evaluate the evidence. (critically appraise the relevant studies you find)
Step 4 of EBP
Determine interventions to use for your particular client(s)
Consider: Quality of the evidence/cl characteristics & context, values & expectations of cl.
Step 5 of EBP
Apply the evidence based intervention.

-Formulate measurable treatment goals with client to aid in eval intervention effectiveness
Step 6 of EBP
Evaluating the intervention used.
Social workers with underdevelopment research skills are at risk for
practicing "primitive, if not incompetent, social work."
SW must be able to distinguish ______ from_____ & be able to eval the _____ & ______of published research.
Good research from bad research
Strengths + weaknesses
Consequences of poor research culture:
Opens the door to a plethora of criticism.
Independent Variable (IV)
influence, affect, or cause the event you want to study (outside causes)

Independent/Intervention
Dependent Variable (DV)
are the key concepts you want to examine. (changes/effects from IV)

Dependent/outcome
Operationalize
To make the independent variable measurable
The scientific method is based on the assumption that
observed effects have a cause.
Basic/Pure Research (Bench Science)
Advanced knowledge with little concern for immediate, practical benefits.
Applied Research
Practical outcome is the focus
Assume some group or society as a whole will gain specific benefits
Hypothesis
A testable prediction, often implied by a theory
Prediction
a guess at the outcome of a hypothesis, not testable
Induction
-involves small samples to predict large
-qualitative research
Deduction
-involves large samples to predict small
-quantitative research (general to more specific)
Exploratory design
Topics with little information available providing inconclusive but point to areas for future research
Descriptive design
Provide descriptive facts about a population (not casual relationship is tested)
Explanatory design
Use of experimental an control groups
Hypothesis testing therefore cause and effect is tested
T/F: Explanatory studies are quantitative studies
True
T/F: Exploratory studies are qualitative studies
True
Operationalize a variable
Translate variables into operational definitions. (observable terms; depression scale)
_____ is key to decrease bias.
Random selection
T/F Studies free from bias have greater generalizability
True
Triangulation
uses a combination of more than one research strategy in a single investigation, cross-check method used to increase the credibility and validity of research findings
Theoretical saturation
Nothing new is being learned
Importance of ethical vs. unethical research
-We can unintentionally harm our subjects.
-Mislead our professional communities.
-Ruin our reputations.
-Cause public backlash.
Deception
research in which the full intent of the study is not explained to potential participants prior to participation.
Nuremburg Code (1947)
A set of ethical guidelines to govern research with human subjects
What is an example of an ethical violation because of deception:
Tuskegee Syphilis Study:
Recruited black men with syphilis & did not provide treatment.

Faden & Beauchamp, 1983:
Injecting cancer cells into 22 geriatric pts without written consent or consulting with primary care.
An example of good research using deception:
Giving participates a placebo, Individuals believe they are receiving a new or effective treatment for their problem.
National Research Act (1974)
Requires approval of an Institutional Review Board (IRB) of any study using human subjects
Belmont Report (1979)
3 Ethical principles of the National Commission for Protection of Human Subjects of Biomedical & Behavioral Research:
-Respect for person: protecting autonomy (self-determination)
-Do not harm & maximize benefit, minimize harm(justice- fair procedures, costs, and benefits)
- Beneficence: Maximizing good outcomes
Information provided by researchers to the IRB about their study's protocol:
-methods & procedures
-risks & benefits
-hypothesis
-recruitment of subjects
3 levels of review:
1. Exempt status
2. Expedited review
3. Full review
Exempt status
-Reviewed by only one or two members, not the entire board
-Poses little or no risk to subjects
-involves the analysis of existing data (secondary data)
T/F: Vulnerable populations cannot be exempted
True.
Examples: elderly, children, disabled, prisoners.
Expedited Review
-for research that is minimal risk
-likelihood & magnitude of harm is not greater than ordinarily encountered in every day life.
-reviewed by a single IRB member or small subcommittee
Full Review
-A thorough review of the research proposal by the full IRB.
-require justification of methodology, deception(if present)
-will examine vulnerable subjects
Informed Consent
Refers to telling potential research participants about all aspects of the research that might reasonably influence their decision to participate.
All research subjects must be volunteers &.... (3 guidelines)
-all must be able to freely decide to participate or not
-must be competent enough to make such a decision
-subjects should be able to withdraw from the study at any time or ask questions
Informed consent must include:
-information of potential risks, benefits, commitments, right to withdraw, procedure, etc.
- contact information for someone, other than on the research team, for whom the subject can discuss any questions, their rights, complaints, etc.
Single system design, N=
Total # of participants in the study
Single system design, n=
subgroup
Baselines (A)
Pre-intervention behavior (give us our starting point for the DV)
Intervention (B)
Independent Variable
Target Behavior
dependent variable
Step 1: Target behavior in SSD should come from the __________.
Client
Step 2: Target behavior in SSD should be ______, __________ or ________.
Counted, observed or quantified.
Step 3: TB is monitored over time, typically involve charting on the
y-axis against time on the x-axis
B Design * Treatment only
Case study does not require a baseline, ex: working with suicidal pts
AB designs
most frequently used by social work practitioners and researchers
internal validity
Measure of whether the treatment was actually responsible for whatever increase is seen. (what was the cause for the change)
external validity
is the measure of whether the results can be generalized to other clients.
T/F: SSD have lower external validity
True
threats to internal validity
-maturation: increase in subjects performance due to natural development
-History: significant special life events that occurs during experimental period
-Testing-Practice Effect: Subjects often do better on repeated testing
Reliability
consistent
Validity
accuracy
Advantages of Single System Design
-do not disrupt the treatment process
-simple
-do not depend upon any particular theoretical framework.
Disadvantages of Single System Design
-Lack of generalizability is frequent
-lack of statistical controls
-cannot effectively control the length of intervention phases.
-reliability & validity are difficult to establish
T/F: SSD will only tell you if something worked for that particular client, not why it worked or why it didn't.
True
Sampling
Gain needed information without experience the entire information w/o experiencing the entire phenomena.

Based on small set of observations can tell about the entire population of interest.

Trends, tendencies, within a population that can be identified with proper sampling.
Population
all of the individuals (unit) that you're interested in
Sample
Subset of a population
probability sampling: Accuracy
External validity, incredibly large populations accurately estimated with small samples
probability sampling: Sampling frame (probability sampling design)
List of members of the population with relevant characteristics. (A student roster; the roster is the sampling frame)
probability sampling: Cross sectional (survey designs)
One time survey with a randomly selected sample, includes everyone.
Simple Random Sample (SRS)
each member of the population and every sample of size n has an equally likely chance of being selected for the sample, researcher assigns a single number to each member
K(sampling interval) =
N/n (Population / sample size)
Sampling interval
the standard distance between elements selected from a population in the sample
Systematic sampling
An efficient alternative to random sampling when K interval is used
Basic ways to determine proper sample size
-Tables
-reputable statistics site
-manually compute
Margin of error
How much your sample results differ from the population
confidence level (reliability)
how likely the results are to be the same if repeated the survey
How much error is acceptable in social sciences?
5% / sig .05
Qualitative research: Sample size is much ______ important for things like margin of error and confidence levels
less
Alternative to probability sampling
When sampling frames cannot be generated. ex: child abuse
Purposive sampling
subjects chosen because they share particular characteristics of interest
Quota sampling
Choosing individual sample members to try and match a known population characteristic as closely as possible.
Longitudinal surveys
trend, cohort, panel
Hallo effect
tendency of an impression to influence.
Positive impression = positive response
Negative impression = negative response
Students also viewed