Evidence-Informed Practice: Research Questions, Search Strategies, and Validity

1. How does evidence-informed practice differ from relying only on research?

EIP integrates research findings with practitioner expertise, client values and preferences, and practice circumstances.

2. What are the main steps of evidence-informed practice?

Formulate an answerable question, search for evidence, critically appraise it, integrate findings into practice, and evaluate outcomes.

3. Why is critical appraisal necessary before applying research?

A published study may have methodological weaknesses or may not apply to the client's circumstances.

4. What distinguishes a research question about effectiveness from one about experience?

Effectiveness asks whether an intervention produces outcomes; experience asks how people perceive or understand a phenomenon.

5. What does PICO stand for?

Population or Problem, Intervention, Comparison, and Outcome.

6. What does PICo stand for in qualitative research?

Population, Interest, and Context.

7. When should PICO be used instead of PICo?

Use PICO for intervention effectiveness and comparisons; use PICo for experiences, meanings, or perspectives.

8. How does an intervention question guide research design?

It usually requires quantitative evidence comparing outcomes, ideally with a credible counterfactual.

9. How does a qualitative question guide research design?

It calls for methods that explore participants' perspectives, experiences, and interpretations.

10. What does AND do in a database search?

Narrows results by requiring both concepts.

11. What does OR do in a database search?

Broadens results by including synonyms or related terms.

12. What does NOT do in a database search?

Excludes a term but may accidentally remove relevant studies.

13. Why group synonyms with OR before connecting concepts with AND?

It captures alternative wording within concepts while requiring the major concepts together.

14. What is the purpose of truncation in searching?

It retrieves multiple word endings using a common root, such as therap* for therapy and therapies.

15. What is the difference between keyword searching and subject headings?

Keywords search authors' wording; subject headings use standardized database indexing terms.

16. Why use more than one database?

Different databases index different journals and disciplines, improving search coverage.

17. What is a research hierarchy used for?

To judge which study designs generally offer stronger evidence for a particular type of question.

18. Why is no single research design best for every question?

Causal questions, lived-experience questions, and descriptive questions require different kinds of evidence.

19. Why are systematic reviews often valued for intervention questions?

They use explicit methods to identify, appraise, and synthesize multiple relevant studies.

20. Why is a systematic review not automatically trustworthy?

Its conclusions depend on search completeness, included-study quality, and synthesis methods.

21. What is internal validity?

The extent to which observed outcomes can reasonably be attributed to the intervention rather than competing explanations.

22. What is external validity?

The extent to which findings can apply to other people, settings, or circumstances.

23. How do internal and external validity differ?

Internal validity concerns causal confidence within the study; external validity concerns applicability beyond it.

24. What is reliability in measurement?

Consistency of a measure across appropriate repeated uses or raters.

25. Why does reliability not guarantee validity?

A measure can consistently produce results while measuring the wrong construct.

26. What three conditions support a causal claim?

The variables are associated, the cause precedes the outcome, and plausible alternative explanations are addressed.

27. What distinguishes a true experiment from a quasi-experiment?

A true experiment uses random assignment to conditions; a quasi-experiment does not.

28. What is the main purpose of random assignment?

To reduce systematic pre-existing differences between intervention and comparison groups.

29. How is random assignment different from random sampling?

Random assignment allocates participants to conditions; random sampling selects participants from a population.

30. Why does random assignment strengthen internal validity?

It makes alternative explanations based on pre-existing group differences less plausible on average.

31. Does random assignment guarantee equivalent groups?

No. Chance imbalances remain possible, especially with small samples.

32. Why is a control or comparison group important?

It helps estimate what might have happened without the intervention.

33. What is the difference between a control group and a treatment group?

The treatment group receives the intervention; the control group receives no intervention, usual care, or another comparison condition.

34. What is a pretest-posttest design?

Outcomes are measured before and after an intervention to examine change.

35. Why is a one-group pretest-posttest design weak for causal claims?

Change could reflect history, maturation, testing, regression, or other causes rather than treatment.

36. What is a posttest-only randomized design?

Participants are randomly assigned and outcomes are measured after the intervention.

37. What is selection bias in a quasi-experiment?

Groups differ before intervention in characteristics that also influence outcomes.

38. Scenario: Motivated clients choose counselling while others decline. Main threat?

Selection bias, because motivation may explain outcome differences.

39. What is history as a threat to internal validity?

An external event during the study affects outcomes independently of the intervention.

40. Scenario: A policy change occurs during an intervention study. Main threat?

History, because the policy could influence outcomes.

41. What is maturation as a validity threat?

Participants change naturally over time, such as through development, aging, or recovery.

42. Scenario: Children improve with age during a six-month program. Main threat?

Maturation, because development could explain improvement.

43. What is testing as a validity threat?

Taking an earlier assessment changes performance on later assessments.

44. What is instrumentation as a validity threat?

Changes in measures, raters, or assessment procedures create apparent outcome differences.

45. What is regression to the mean?

Extreme initial scores tend to be less extreme on later measurement even without intervention.

46. Scenario: Clients selected for unusually high distress later score lower. Possible threat?

Regression to the mean, especially without a comparison group.

47. What is attrition or experimental mortality?

Participants drop out, potentially biasing findings if dropout differs across groups or relates to outcomes.

48. Scenario: More severely affected clients leave the treatment group. Main concern?

Attrition bias may make the intervention appear more effective than it is.

49. What is treatment contamination or diffusion?

Comparison participants receive elements of the intervention, reducing separation between groups.

50. What is treatment fidelity?

The extent to which an intervention is delivered as intended.

51. Why might social workers choose quasi-experimental designs?

Random assignment may be infeasible or unethical in real-world service settings.

52. What is a nonequivalent comparison-group design?

An intervention group is compared with a nonrandomly formed group.

53. How can baseline comparisons strengthen a quasi-experiment?

They reveal initial group differences that may complicate interpretation.

54. What is an interrupted time-series design?

Outcomes are measured repeatedly before and after an intervention or event.

55. Why are multiple pre-intervention observations valuable in time-series research?

They reveal baseline trends and variability, helping distinguish intervention effects from ongoing change.

56. Scenario: Monthly hospitalizations are tracked for a year before and after a policy. Design?

Interrupted time-series design.

57. What is a single-case experimental design?

Repeated observations examine intervention effects for one client or a small number of cases.

58. What does A-B mean in single-case design?

A is baseline measurement; B is measurement during intervention.

59. What is a major weakness of an A-B design?

Without replication or withdrawal, changes may have causes other than the intervention.

60. What does A-B-A-B add to a single-case design?

Withdrawal and reintroduction of intervention provide repeated evidence of an intervention-outcome relationship.

61. When might withdrawal designs be ethically inappropriate?

When removing an effective intervention could cause harm or withhold needed support.

62. What is the purpose of a multiple-baseline design?

To introduce intervention at different times across people, settings, or behaviours without withdrawing it.

63. What is the central purpose of qualitative research?

To understand experiences, meanings, perspectives, and processes in context.

64. How does purposive sampling differ from probability sampling?

Purposive sampling selects information-rich participants; probability sampling uses known selection probabilities.

65. What is credibility in qualitative appraisal?

Whether interpretations are believable and well supported by participants' accounts and the research process.

66. What is transferability in qualitative research?

Whether sufficiently described findings may be relevant in other contexts.

67. What is dependability in qualitative research?

Whether the research process is logical, documented, and consistent with the study approach.

68. What is confirmability in qualitative research?

Whether interpretations are grounded in data rather than primarily researcher assumptions.

69. What is reflexivity in qualitative research?

Researchers critically examine how their positions, assumptions, and relationships shape the study.

70. How can triangulation strengthen qualitative research?

Comparing sources, methods, or perspectives can deepen understanding and examine consistency.

71. What is member checking?

Inviting participants to review or comment on data or interpretations, when appropriate.

72. Why is prolonged engagement relevant to qualitative credibility?

Sustained involvement may build contextual understanding and trust, though it is not required for every qualitative design.

73. What is thematic analysis?

Systematically identifying and interpreting patterns of meaning across qualitative data.

74. Scenario: Interviews describe fatigue but include little recruitment detail. Main appraisal concern?

Limited recruitment transparency makes it harder to judge sample relevance and possible selection limitations.

75. How should you answer a scenario-based research-design question?

Identify the question's purpose, locate design clues, name the relevant strength or threat, and explain how it affects the conclusion.